22 |
23 | Overview
24 | Download
25 | Cite
26 | Open Resources
27 | issues
28 |
29 |
30 |
31 |
32 | {{ content }}
33 |
34 |
39 |
40 |
41 | {% if site.google_analytics %}
42 |
51 | {% endif %}
52 |
53 |
54 |
59 |
60 |
61 |
--------------------------------------------------------------------------------
/qc.md:
--------------------------------------------------------------------------------
1 | ---
2 | ---
3 |
4 | ## Quality Control/Assurance (QC)
5 |
6 | QC is performed at various stages of the dHCP analysis as listed below. The
7 | final selection of subject data to release made use of these as outlined
8 | subsequently.
9 |
10 | 1. After **reconstruction** the T2, T1, T13D, fMRI and dMRI datasets were visually
11 | inspected and scored as PASS/FAIL.
12 |
13 | 2. The **fMRI pipeline** and **dMRI pipelines** generate numerous quantitative
14 | QC metrics which are described on their respective pages:
15 |
16 | 1. [Structural Pipeline](struct.md#struct-qc)
17 |
18 | 2. [Functional Pipeline](fmri.md#fmri-qc)
19 |
20 | 2. [Diffusion EDDY Pipeline](dwi.md#eddy-qc)
21 |
22 | 3. [Diffusion SHARD Pipeline](dwi-shard.md#shard-qc)
23 |
24 | 3. All QC metrics are available in the `combined.tsv` spreadsheet in the
25 | [supplementary](https://github.com/BioMedIA/dHCP-release-notes/tree/master/supplementary_files).
26 |
27 | ## Selection Criteria
28 |
29 | The inclusion criteria for **reconstructed/raw data** is:
30 |
31 | 1. Scans that scored PASS on the visual recon QC are released, and also
32 | subsequently processed by the sMRI, fMRI, and dMRI pipeline.
33 |
34 | The inclusion criteria for the **sMRI pipeline outputs** is:
35 |
36 | 1. Raw T2w must pass the recon QC.
37 |
38 | 2. All completed segmentations from the structural pipeline are released.
39 |
40 | 3. Extracted surfaces were visually inspected and only successful extractions
41 | are released.
42 |
43 | 4. Myelin maps are only released if the surface extraction was successful.
44 |
45 | The inclusion criteria for **fMRI pipeline outputs** is:
46 |
47 | 1. Raw fMRI & T2w must pass the recon QC.
48 |
49 | 2. Preprocessed T2w must pass the sMRI pipeline QC.
50 |
51 | 3. All fMRI that meets the inclusion criteria is processed and released. This includes data that subsequently failed the fMRI pipeline QC. Data that failed fMRI
52 | pipeline QC is flagged in the `combined.tsv` spreadsheet in the
53 | [supplementary](https://github.com/BioMedIA/dHCP-release-notes/tree/master/supplementary_files)
54 |
55 | 4. Transforms to standard space are dependent upon availability of extracted
56 | (white and pial) surfaces from sMRI pipeline.
57 |
58 | The inclusion criteria for **dMRI pipeline outputs** is:
59 |
60 | 1. Raw dMRI & T2w must pass the recon QC.
61 |
62 | 2. All preprocessed dMRI is released. Data that failed dMRI
63 | pipeline/s QC is flagged in the `combined.tsv` spreadsheet in the
64 | [supplementary](https://github.com/BioMedIA/dHCP-release-notes/tree/master/supplementary_files)
65 |
66 | 3. Transforms to standard space are dependent upon availability of extracted
67 | (white and pial) surfaces from sMRI pipeline.
68 |
69 |
--------------------------------------------------------------------------------
/index.md:
--------------------------------------------------------------------------------
1 | ---
2 | ---
3 |
4 | # Consolidated release, now including data from both fetuses and neonates
5 |
6 | Details can be found in the neonatal data release paper [The Developing Human Connectome Project Neonatal Data Release](https://pubmed.ncbi.nlm.nih.gov/35677357/).
7 |
8 | Fetal data release paper in preparation.
9 |
10 |
11 | ## 1. Cohort
12 | • Neonatal participants: 783 subjects (886 datasets).
13 | • Fetal participants: 273 subjects (297 datasets).
14 |
15 | ## 2. MR Imaging data
16 | • Native and processed. Structural, DTI, fMRI.
17 |
18 | ## 3. Clinical information about pregnancy and birth
19 | • Pregnancy: Data include previous obstetric history, pre-pregnancy and antenatal maternal conditions, medication.
20 |
21 | • Birth: data include gestational age at birth, birth weight, length, and head circumference, presentation and mode of birth, medication required at delivery, nutrition and feeding, apgar scores at 1 and 5 min of age. For babies that were admitted to the intensive care unit, there are clinical details about their stay.
22 |
23 | ## 4. Genomics
24 | • Saliva-derived DNA samples were genotyped for single-nucleotide polymorphisms and also processed for methylation analysis.
25 |
26 | ## 5. Eye-tracking data at 18 months of age
27 | • Native data collected during visual attention tasks.
28 |
29 | • Processed measures of social, non-social, exogenous and endogenous attention.
30 |
31 | ## 6. Neurodevelopmental outcome at 18 months of age
32 | • Bayley-III Scales of Infant Development.
33 | • Neurological examination.
34 | • Parenal completed questionnaires about child's behaviour, parenting style and home environment.
35 |
36 |
37 |
38 | ---
39 |
40 |
41 |
42 |
43 | ## HOW TO ACCESS AND DOWNLOAD THE DATA:
44 | Instructions on how to download data can be found [here](https://biomedia.github.io/dHCP-release-notes/supplementary_files/Guidelines%20downloading%20data%20v3.pdf).
45 |
46 |
47 |
48 |
61 |
62 | ## Acknowledgments
63 |
64 | The research leading to these data has received funding from the European
65 | Research Council under the European Union Seventh Framework Programme
66 | (FP/20072013)/ERC Grant Agreement no. 319456. The work was also supported
67 | by the NIHR Biomedical Research Centres at Guys and St Thomas NHS Trust.
68 | We are grateful to the families who generously supported this trial.
69 | We would like to acknowledge Core support for data acquisition was provided
70 | by the Wellcome/EPSRC Centre for Medical Engineering [WT 203148/Z/16/Z]. We are
71 | also thankful to the WU-Minn-Oxford Human Connectome Project consortium
72 | (1U54MH091657-01) for access to their computing resources.
73 |
74 |
75 |
76 |
77 |
--------------------------------------------------------------------------------
/release3_notes.md:
--------------------------------------------------------------------------------
1 | ---
2 | ---
3 | ## Third data release (2023)
4 | ## Consolidated release, now including data from both fetuses and neonates
5 |
6 | ## 1. Cohort
7 | • Neonatal participants: 783 subjects (886 datasets).
8 | • Fetal participants: 273 subjects (297 datasets).
9 |
10 | ## 2. MR Imaging data
11 | • Native and processed. Structural, DTI, fMRI.
12 |
13 | ## 3. Clinical information about pregnancy and birth
14 | • Pregnancy: Data include previous obstetric history, pre-pregnancy and antenatal maternal conditions, medication.
15 | • Birth: data include gestational age at birth, birth weight, length, and head circumference, presentation and mode of birth, medication required at delivery, nutrition and feeding, apgar scores at 1 and 5 min of age. For babies that were admitted to the intensive care unit, there are clinical details about their stay.
16 |
17 | ## 4. Genomics
18 | • Saliva-derived DNA sample were genotyped for single-nucleotide polymorphisms and also processed for methylation analysis.
19 |
20 | ## 5. Eye-tracking data at 18 months of age
21 | • Visual attention metrics
22 |
23 | ## Upcoming data (pending NDA approval):
24 | • Neurodevelopmental outcome at 18 months of age.
25 |
26 |
27 | ## HOW TO ACCESS AND DOWNLOAD THE DATA:
28 | All data are available from the NDA website [NDA website](https://nda.nih.gov/edit_collection.html?id=3955). The NDA is responsible for all data governance. Instructions for how to access and download the data can be found:
29 |
30 | [Updated guidelines - download from NDA](https://github.com/BioMedIA/dHCP-release-notes/tree/master/supplementary_files/Guidelines%20downloading%20data%20v3.pdf)
31 |
32 |
33 | ## Feedback
34 | We invite colleagues in the field to explore and
35 | [feedback](https://neurostars.org/tags/developing-hcp) on the value and
36 | characteristics of the image dataset. The image data have been processed
37 | using analysis pipelines that are subject to further development. If you
38 | use this data or the pipelines please cite the appropriate publications as
39 | detailed in the [How to cite](cite.html) notes.
40 |
41 |
42 |
43 |
56 |
57 | ## Acknowledgments
58 |
59 | The research leading to these data has received funding from the European
60 | Research Council under the European Union Seventh Framework Programme
61 | (FP/20072013)/ERC Grant Agreement no. 319456. The work was also supported
62 | by the NIHR Biomedical Research Centres at Guys and St Thomas NHS Trust.
63 | We are grateful to the families who generously supported this trial.
64 | We would like to acknowledge Core support for data acquisition was provided
65 | by the Wellcome/EPSRC Centre for Medical Engineering [WT 203148/Z/16/Z]. We are
66 | also thankful to the WU-Minn-Oxford Human Connectome Project consortium
67 | (1U54MH091657-01) for access to their computing resources.
68 |
69 |
70 |
71 |
72 |
--------------------------------------------------------------------------------
/assets/javascript/toc.js:
--------------------------------------------------------------------------------
1 | // https://github.com/ghiculescu/jekyll-table-of-contents
2 | (function($){
3 | $.fn.toc = function(options) {
4 | var defaults = {
5 | noBackToTopLinks: false,
6 | title: 'Jump to...',
7 | minimumHeaders: 3,
8 | headers: 'h1, h2, h3, h4, h5, h6',
9 | listType: 'ol', // values: [ol|ul]
10 | showEffect: 'show', // values: [show|slideDown|fadeIn|none]
11 | showSpeed: 'slow', // set to 0 to deactivate effect
12 | classes: { list: '',
13 | item: '',
14 | link: ''
15 | }
16 | },
17 | settings = $.extend(defaults, options);
18 |
19 | function fixedEncodeURIComponent (str) {
20 | return encodeURIComponent(str).replace(/[!'()*]/g, function(c) {
21 | return '%' + c.charCodeAt(0).toString(16);
22 | });
23 | }
24 |
25 | function createLink (header) {
26 | var innerText = (header.textContent === undefined) ? header.innerText : header.textContent;
27 | return "" + innerText + "";
28 | }
29 |
30 | var headers = $(settings.headers).filter(function() {
31 | // get all headers with an ID
32 | var previousSiblingName = $(this).prev().attr( "name" );
33 | if (!this.id && previousSiblingName) {
34 | this.id = $(this).attr( "id", previousSiblingName.replace(/\./g, "-") );
35 | }
36 | return this.id;
37 | }), output = $(this);
38 | if (!headers.length || headers.length < settings.minimumHeaders || !output.length) {
39 | $(this).hide();
40 | return;
41 | }
42 |
43 | if (0 === settings.showSpeed) {
44 | settings.showEffect = 'none';
45 | }
46 |
47 | var render = {
48 | show: function() { output.hide().html(html).show(settings.showSpeed); },
49 | slideDown: function() { output.hide().html(html).slideDown(settings.showSpeed); },
50 | fadeIn: function() { output.hide().html(html).fadeIn(settings.showSpeed); },
51 | none: function() { output.html(html); }
52 | };
53 |
54 | var get_level = function(ele) { return parseInt(ele.nodeName.replace("H", ""), 10); };
55 | var highest_level = headers.map(function(_, ele) { return get_level(ele); }).get().sort()[0];
56 | var return_to_top = '';
57 |
58 | var level = get_level(headers[0]),
59 | this_level,
60 | html = settings.title + " <" +settings.listType + " class=\"" + settings.classes.list +"\">";
61 | headers.on('click', function() {
62 | if (!settings.noBackToTopLinks) {
63 | window.location.hash = this.id;
64 | }
65 | })
66 | .addClass('clickable-header')
67 | .each(function(_, header) {
68 | this_level = get_level(header);
69 | if (!settings.noBackToTopLinks && this_level === highest_level) {
70 | $(header).addClass('top-level-header').after(return_to_top);
71 | }
72 | if (this_level === level) // same level as before; same indenting
73 | html += "
" + createLink(header);
74 | else if (this_level <= level){ // higher level than before; end parent ol
75 | for(var i = this_level; i < level; i++) {
76 | html += "
"
77 | }
78 | html += "
" + createLink(header);
79 | }
80 | else if (this_level > level) { // lower level than before; expand the previous to contain a ol
81 | for(i = this_level; i > level; i--) {
82 | html += "<" + settings.listType + " class=\"" + settings.classes.list +"\">" +
83 | "
"
84 | }
85 | html += createLink(header);
86 | }
87 | level = this_level; // update for the next one
88 | });
89 | html += "";
90 | if (!settings.noBackToTopLinks) {
91 | $(document).on('click', '.back-to-top', function() {
92 | $(window).scrollTop(0);
93 | window.location.hash = '';
94 | });
95 | }
96 |
97 | render[settings.showEffect]();
98 | };
99 | })(jQuery);
100 |
--------------------------------------------------------------------------------
/LICENSE:
--------------------------------------------------------------------------------
1 | Creative Commons Legal Code
2 |
3 | CC0 1.0 Universal
4 |
5 | CREATIVE COMMONS CORPORATION IS NOT A LAW FIRM AND DOES NOT PROVIDE
6 | LEGAL SERVICES. DISTRIBUTION OF THIS DOCUMENT DOES NOT CREATE AN
7 | ATTORNEY-CLIENT RELATIONSHIP. CREATIVE COMMONS PROVIDES THIS
8 | INFORMATION ON AN "AS-IS" BASIS. CREATIVE COMMONS MAKES NO WARRANTIES
9 | REGARDING THE USE OF THIS DOCUMENT OR THE INFORMATION OR WORKS
10 | PROVIDED HEREUNDER, AND DISCLAIMS LIABILITY FOR DAMAGES RESULTING FROM
11 | THE USE OF THIS DOCUMENT OR THE INFORMATION OR WORKS PROVIDED
12 | HEREUNDER.
13 |
14 | Statement of Purpose
15 |
16 | The laws of most jurisdictions throughout the world automatically confer
17 | exclusive Copyright and Related Rights (defined below) upon the creator
18 | and subsequent owner(s) (each and all, an "owner") of an original work of
19 | authorship and/or a database (each, a "Work").
20 |
21 | Certain owners wish to permanently relinquish those rights to a Work for
22 | the purpose of contributing to a commons of creative, cultural and
23 | scientific works ("Commons") that the public can reliably and without fear
24 | of later claims of infringement build upon, modify, incorporate in other
25 | works, reuse and redistribute as freely as possible in any form whatsoever
26 | and for any purposes, including without limitation commercial purposes.
27 | These owners may contribute to the Commons to promote the ideal of a free
28 | culture and the further production of creative, cultural and scientific
29 | works, or to gain reputation or greater distribution for their Work in
30 | part through the use and efforts of others.
31 |
32 | For these and/or other purposes and motivations, and without any
33 | expectation of additional consideration or compensation, the person
34 | associating CC0 with a Work (the "Affirmer"), to the extent that he or she
35 | is an owner of Copyright and Related Rights in the Work, voluntarily
36 | elects to apply CC0 to the Work and publicly distribute the Work under its
37 | terms, with knowledge of his or her Copyright and Related Rights in the
38 | Work and the meaning and intended legal effect of CC0 on those rights.
39 |
40 | 1. Copyright and Related Rights. A Work made available under CC0 may be
41 | protected by copyright and related or neighboring rights ("Copyright and
42 | Related Rights"). Copyright and Related Rights include, but are not
43 | limited to, the following:
44 |
45 | i. the right to reproduce, adapt, distribute, perform, display,
46 | communicate, and translate a Work;
47 | ii. moral rights retained by the original author(s) and/or performer(s);
48 | iii. publicity and privacy rights pertaining to a person's image or
49 | likeness depicted in a Work;
50 | iv. rights protecting against unfair competition in regards to a Work,
51 | subject to the limitations in paragraph 4(a), below;
52 | v. rights protecting the extraction, dissemination, use and reuse of data
53 | in a Work;
54 | vi. database rights (such as those arising under Directive 96/9/EC of the
55 | European Parliament and of the Council of 11 March 1996 on the legal
56 | protection of databases, and under any national implementation
57 | thereof, including any amended or successor version of such
58 | directive); and
59 | vii. other similar, equivalent or corresponding rights throughout the
60 | world based on applicable law or treaty, and any national
61 | implementations thereof.
62 |
63 | 2. Waiver. To the greatest extent permitted by, but not in contravention
64 | of, applicable law, Affirmer hereby overtly, fully, permanently,
65 | irrevocably and unconditionally waives, abandons, and surrenders all of
66 | Affirmer's Copyright and Related Rights and associated claims and causes
67 | of action, whether now known or unknown (including existing as well as
68 | future claims and causes of action), in the Work (i) in all territories
69 | worldwide, (ii) for the maximum duration provided by applicable law or
70 | treaty (including future time extensions), (iii) in any current or future
71 | medium and for any number of copies, and (iv) for any purpose whatsoever,
72 | including without limitation commercial, advertising or promotional
73 | purposes (the "Waiver"). Affirmer makes the Waiver for the benefit of each
74 | member of the public at large and to the detriment of Affirmer's heirs and
75 | successors, fully intending that such Waiver shall not be subject to
76 | revocation, rescission, cancellation, termination, or any other legal or
77 | equitable action to disrupt the quiet enjoyment of the Work by the public
78 | as contemplated by Affirmer's express Statement of Purpose.
79 |
80 | 3. Public License Fallback. Should any part of the Waiver for any reason
81 | be judged legally invalid or ineffective under applicable law, then the
82 | Waiver shall be preserved to the maximum extent permitted taking into
83 | account Affirmer's express Statement of Purpose. In addition, to the
84 | extent the Waiver is so judged Affirmer hereby grants to each affected
85 | person a royalty-free, non transferable, non sublicensable, non exclusive,
86 | irrevocable and unconditional license to exercise Affirmer's Copyright and
87 | Related Rights in the Work (i) in all territories worldwide, (ii) for the
88 | maximum duration provided by applicable law or treaty (including future
89 | time extensions), (iii) in any current or future medium and for any number
90 | of copies, and (iv) for any purpose whatsoever, including without
91 | limitation commercial, advertising or promotional purposes (the
92 | "License"). The License shall be deemed effective as of the date CC0 was
93 | applied by Affirmer to the Work. Should any part of the License for any
94 | reason be judged legally invalid or ineffective under applicable law, such
95 | partial invalidity or ineffectiveness shall not invalidate the remainder
96 | of the License, and in such case Affirmer hereby affirms that he or she
97 | will not (i) exercise any of his or her remaining Copyright and Related
98 | Rights in the Work or (ii) assert any associated claims and causes of
99 | action with respect to the Work, in either case contrary to Affirmer's
100 | express Statement of Purpose.
101 |
102 | 4. Limitations and Disclaimers.
103 |
104 | a. No trademark or patent rights held by Affirmer are waived, abandoned,
105 | surrendered, licensed or otherwise affected by this document.
106 | b. Affirmer offers the Work as-is and makes no representations or
107 | warranties of any kind concerning the Work, express, implied,
108 | statutory or otherwise, including without limitation warranties of
109 | title, merchantability, fitness for a particular purpose, non
110 | infringement, or the absence of latent or other defects, accuracy, or
111 | the present or absence of errors, whether or not discoverable, all to
112 | the greatest extent permissible under applicable law.
113 | c. Affirmer disclaims responsibility for clearing rights of other persons
114 | that may apply to the Work or any use thereof, including without
115 | limitation any person's Copyright and Related Rights in the Work.
116 | Further, Affirmer disclaims responsibility for obtaining any necessary
117 | consents, permissions or other rights required for any use of the
118 | Work.
119 | d. Affirmer understands and acknowledges that Creative Commons is not a
120 | party to this document and has no duty or obligation with respect to
121 | this CC0 or use of the Work.
122 |
--------------------------------------------------------------------------------
/acquire.md:
--------------------------------------------------------------------------------
1 | ---
2 | ---
3 |
4 | ## Subjects
5 |
6 | Infants were recruited and imaged at the Evelina Newborn Imaging Centre,
7 | St Thomas' Hospital, London, UK. The study was approved by the UK Health
8 | Research Authority (Research Ethics Committee reference number: 14/LO/1169)
9 | and written parental consent was obtained in every case for imaging and data
10 | release. The images included in this release were obtained from 783 subjects
11 | born between 23-44 gestational weeks: 578 born and scanned at term-equivalent
12 | age which we defined as 37-44 weeks, 156 preterm-born subjects scanned soon
13 | after birth and 133 preterm-born subjects scanned at term-equivalent age.
14 | The images have been reviewed for evidence of anomalies and abnormalities
15 | and a radiology score is provided, although users should verify that data
16 | they use are fit for their purposes.
17 |
18 | ## Overview of data
19 |
20 | The data release contains structural (T1w and T2w) resting state functional
21 | and diffusion images supplied as original image data and after preprocessing
22 | pipelines as described below have been applied. The neonatal brain has
23 | different tissue properties to adult brain, most strikingly it has a
24 | higher water content and myelination of white matter is incomplete.
25 | Consequently, the relaxation times T1 and T2 are longer than in adult brain
26 | and white matter has longer T1 and T2 than grey matter. In neonates, brain
27 | anatomy is revealed more clearly in T2w than T1w images and thus the former
28 | are treated as the primary data for anatomical segmentation and to provide
29 | the anatomical substrates needed for functional and diffusion analysis. All
30 | neonates (with 6 exceptions) were imaged in natural sleep. If the baby woke
31 | up scanning was halted and attempts made to re-settle the subject without
32 | taking them out of the patient immobilization system. Even when sleeping
33 | peacefully, many babies move and so all data were motion corrected, mostly
34 | using methods developed specifically for the dHCP project. As a result of
35 | the challenges of imaging unsedated infants we were not able to obtain high
36 | quality and complete data for every modality on every subject.
37 |
38 | There were 886 sessions with T2w images that passed QC and an additional
39 | session without T2w images. Of those 887 sessions, 818 had fMRI data that
40 | passed QC and 758 had dMRI data that passed QC. The T1w images were not
41 | subject to the same level of systematic QC as they were not processed by
42 | pre-processing pipelines. Because of their lower anatomical importance, the
43 | T1w images were placed at the end of the protocol and are of more variable
44 | quality than the T2w data. The release contains 711 sessions with T1w
45 | multi-slice fast spin-echo images and 734 sessions with T1 MPRAGE images.
46 |
47 | There is a spread of gestational ages with 578 subjects in the term equivalent
48 | age range, which we defined as 37 to 44 gestational weeks. Also, although
49 | these subjects were recruited as "normal subjects" (with clearly specified
50 | inclusion and exclusion criteria), there were inevitably incidental findings
51 | on the images obtained. All the anatomical images were reviewed by an expert
52 | perinatal neuroradiologist who scored the subjects using a 5-point scale
53 | (see below) -- this information is provided
54 |
55 | ## Acquisition details
56 |
57 | Imaging was carried out on 3T Philips Achieva (running modified R3.2.2
58 | software) using a dedicated neonatal imaging system which included a neonatal
59 | 32 channel phased array head coil1. Infants were imaged without
60 | sedation except for 6 who are indicated. Anatomical images (T1w and T2w),
61 | resting state functional (rs-fMRI) and diffusion (dMRI) acquisitions were
62 | acquired in a total examination time of 63 minutes. Sequence parameters
63 | were as follows:
64 |
65 | **Calibration scans:** B0 mapping was performed using an interleaved dual TE
66 | spoiled gradient echo sequence and localised image-based shimming performed
67 | for use with all EPI sequences as described in2. B0 field maps
68 | using the optimised higher order shims were subsequently re-acquired between
69 | the fMRI and dMRI acquisitions. B1 mapping was performed using the dual
70 | refocusing echo acquisition mode (DREAM)3 method, with STE first
71 | and STEAM flip angle of 60.
72 |
73 | **Anatomical acquisition:** T2w and inversion recovery T1w multi-slice fast
74 | spin-echo images were each acquired in sagittal and axial slice stacks with
75 | in-plane resolution 0.8x0.8mm2 and 1.6mm slices overlapped by 0.8mm (except in
76 | T1w Sagittal which used a slice overlap of 0.74mm). Other parameters were –
77 | T2w: 12000/156ms TR/TE, SENSE factor 2.11 (axial) and 2.60 (sagittal); T1w:
78 | 4795/1740/8.7ms TR/TI/TE, SENSE factor 2.27 (axial) and 2.66 (sagittal).
79 | In addition, 3D MPRAGE was acquired with 0.8mm isotropic resolution and
80 | parameters: 11/4.6/1400ms TR/TE/TI, SENSE factor 1.2 (RL).
81 |
82 | **rs-fMRI:** High temporal resolution fMRI developed for neonates4
83 | used multiband (MB) 9x accelerated echo-planar imaging and was collected for
84 | 15 minutes, TE/TR=38/392ms gave 2300 volumes, with an acquired resolution
85 | of 2.15mm isotropic. No in-plane acceleration or partial Fourier was
86 | used. Single-band reference scans were also acquired with bandwidth matched
87 | readout, along with additional spin-echo acquisitions with both AP/PA
88 | fold-over encoding directions.
89 |
90 | Physiological recordings of VCG, PPU and respiratory traces are provided
91 | unprocessed in the sourcedata folder. Alignment to rs-fMRI data can be achieved
92 | by means of locating the 'end of scan' marker (scripts are available
93 | to aid loading and interpretation of this file5), and knowing the
94 | frequency of the recordings (496Hz) and TR x number of volumes acquired
95 | (0.392s x 2300) in order to identify the start of scan point.
96 | Note, for improved accuracy on this cohort a small delay of ~85ms between the
97 | true end of data acquisition and 'end of scan' marker has been identified,
98 | after accounting for this the precision of identifying the true start of scan
99 | in the physiological file should be of the order +/- 50ms, for a complete scan of 15 minutes duration.
100 |
101 | **dMRI:** A spherically optimized set of directions on 4 shells (b0:
102 | 20, b400: 64, b1000: 88, b2600: 128)6 was split into 4 optimal
103 | subsets (one per Phase Encoding Direction). These directions were then spread
104 | temporally taking motion and duty cycle considerations into account. If the
105 | baby woke up during the diffusion scan, the acquisition could be halted
106 | and restarted (after resettling the subject) with a user defined overlap
107 | in acquired diffusion weightings7. Acceleration of MB 4, SENSE
108 | factor 1.2 and Partial Fourier 0.86 was used, acquired resolution 1.5x1.5mm,
109 | 3mm slices with 1.5mm overlap, 3800/90ms TR/TE.
110 |
111 | ## References
112 |
113 | 1. Hughes, E. J., Winchman, T., Padormo, F., Teixeira, R., Wurie, J.,
114 | Sharma, M., Fox, M, Hutter, J., Cordero-Grande, L., Price, A. N., Allsop,
115 | J,, Bueno-Conde, J., Tusor, N.,, Arichi, T., Edwards, A. D., Rutherford,
116 | M. A., Counsell, S. J., and Hajnal J. V. **A dedicated neonatal brain
117 | imaging system** *Magnetic Resonance in Medicine (2017), 78: 794-804.*
118 | [DOI: 10.1002/mrm.26462](https://doi.org/10.1002/mrm.26462)
119 |
120 | 2. Gaspar, A. [**Improving foetal and neonatal echo-planar imaging with
121 | image-based shimming**](https://repositorio.ul.pt/handle/10451/22886)
122 | *Master Thesis (2015), Universidade de Lisboa.*
123 |
124 | 3. Nehrke, K. and Börnert, P. **DREAM—a novel approach for robust, ultrafast, multislice B1 mapping.**
125 | *Magnetic Resonance in Medicine (2012), 68: 1517-1526.* [DOI: 10.1002/mrm.24158](https://doi.org/10.1002/mrm.24158)
126 |
127 | 4. Price, A.N., Cordero-Grande, L., Malik, S.J.,
128 | Abaei, M., Arichi, T., Hughes, E., Rueckert, D.,
129 | Edwards, A.D., Hajnal, J.V. [**Accelerated neonatal fMRI using multiband
130 | EPI**](http://www.developingconnectome.org/wp-content/uploads/sites/70/2019/08/Accelerated-Neonatal-fMRI-using-Multiband-EPI.-ISMRM-2015.pdf)
131 | *ISMRM 2015: 3911.*
132 |
133 | 5. [**ReadPhilipsScanPhysLog.m** by Paul Groot (MathWorks File Exchange)](https://uk.mathworks.com/matlabcentral/fileexchange/42100-readphilipsscanphyslog-filename-channels-skipprep)
134 |
135 | 6. Tournier, J.D., Hughes, E., Turso, N., Sotiropoulos,
136 | N. S., Jbadhi, S., Andersson, J., Reuckert, D., Edwards,
137 | A. D., Hajnal, J. V. [**Data-driven optimisation of multi-shell
138 | HARDI**](http://www.developingconnectome.org/wp-content/uploads/sites/70/2019/08/Data-driven-optimisation-of-multi-shell-HARDI.pdf)
139 | *ISMRM 2015: 2897.*
140 |
141 | 7. Hutter, J., Tournier J. D., Price, A. N., Cordero-Grande, L., Hughes,
142 | E. J., Malik, S., Steinweg, J., Bastiani, M., Sotiropoulos, S. N., Jbabdi,
143 | S., Andersson, J., Edwards, A. D., and Hajnal, J. V. **Time-efficient
144 | and flexible design of optimised multi-shell HARDI diffusion** *Magnetic
145 | Resonance in Medicine (2018), 79: 1276-1292.* [DOI: 10.1002/mrm.26765
146 | ](https://doi.org/10.1002/mrm.26765)
147 |
148 |
--------------------------------------------------------------------------------
/dwi.md:
--------------------------------------------------------------------------------
1 | ---
2 | ---
3 |
4 | ## Diffusion EDDY Pipeline
5 |
6 | ### Inputs
7 |
8 | **From reconstruction pipeline:** `rawdata/sub-{subid}/ses-{sesid}`
9 |
10 | | Description | Filename |
11 | |:----------------------------------------------------------------------------------------|:--------------------------------------------------------------------------|
12 | | Multi-band dMRI EPI (Release 2 reconstruction) | `dwi/sub-{subid}_ses-{sesid}_rec-release2_dwi.nii` |
13 |
14 | ### Outputs
15 |
16 | The primary outputs of the diffusion EDDY preprocessing pipeline are:
17 |
18 | **Path:** `derivatives/dhcp_dmri_eddy_pipeline/sub-{subid}/ses-{sesid}`
19 |
20 | | Description | Filename |
21 | |:---------------------------------------------------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------|
22 | | Eddy current, susceptibility-by-motion and motion (within and between volumes) corrected super-resolved 4D volume with outlier rejection and replacement | `dwi/sub-{subid}_ses-{sesid}_desc-preproc_dwi.nii.gz` |
23 | | List of b-values | `dwi/sub-{subid}_ses-{sesid}_desc-preproc_dwi.bval` |
24 | | List of post-EDDY rotated gradient directions | `dwi/sub-{subid}_ses-{sesid}_desc-preproc_dwi.bvec` |
25 | | Warp from diffusion space to the extended dHCP 40-week template space | `xfm/sub-{subid}_ses-{sesid}_from-dwi_to-extdhcp40wk_mode-image.nii.gz` |
26 |
27 | The complete list of outputs and QC reports generated by the diffusion EDDY pipeline
28 | are listed in the [Diffusion EDDY pipeline](structure.html#diffusion-eddy-pipeline) section of the directory structure summary.
29 |
30 | ### Pipeline
31 |
32 | For a complete and detailed description of all the steps
33 | involved in the [dHCP neonatal diffusion MRI (dMRI) data processing
34 | pipeline](https://git.fmrib.ox.ac.uk/matteob/dHCP_neo_dMRI_pipeline_release),
35 | the reader is referred elsewhere1. The main processing steps
36 | are briefly summarised below:
37 |
38 | 1. For each phase encoding (PE) direction, the diffusion un-weighted
39 | b0 volume pairs least-affected by intra-volume motion are automatically
40 | selected. The dataset is then re-organised by moving the least-affected
41 | b0 volume and the volumes that follow (until the end of the acquisition)
42 | at the beginning of the 4D raw data file.
43 |
44 | 2. Field maps for correcting susceptibility-induced distortions are estimated
45 | using FSL TOPUP2.
46 |
47 | 3. Distortions caused by susceptibility, between-volume motion, within-volum
48 | motion, motion-induced signal drop-out, motion-by-susceptibility interactions,
49 | and eddy currents are corrected; outlier slices are detected and replaced
50 | in raw distorted space. All these steps use FSL EDDY3-6.
51 |
52 | 4. A super-resolution algorithm7 is applied along the
53 | slice-selection direction, to achieve isotropic resolution of 1.5 mm.
54 |
55 | 5. Post-processing using traditional tensor fitting, as well as FSL's
56 | [BedpostX](https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FDT/UserGuide) for
57 | multishell data13 is applied.
58 |
59 | 6. Diffusion data are aligned to high-resolution structural (T2-weighted)
60 | space using boundary-based registration8,9 on the average
61 | attenuation volume for the b=1000 s/mm2 shell (i.e. b1k/b0). This
62 | transformation is combined with a non-linear registration10 of
63 | the T2w volume to the 40 weeks template11 to allow transformations
64 | between diffusion and atlas spaces.
65 |
66 |
67 | ## Diffusion MRI QC
68 |
69 | 1. Numerous quality assurance metrics are calculated by the EDDY QC
70 | tools12. Four of these are specifically compared against
71 | the population distribution to flag outliers for manual inspection and
72 | potential exclusion:
73 |
74 | 1. Mean signal-to-noise ratio (SNR) from the b0 volumes
75 |
76 | 2. Mean contrast-to-noise ratio (CNR) for each b-shell, i.e., 400,
77 | 1000 and 2600 s/mm2.
78 |
79 | 2. All QC metrics are then converted to Z-scores and averaged, to generate
80 | a summary QC metric.
81 |
82 | 3. All QC metrics are available in the `combined.tsv` spreadsheet in the
83 | [supplementary](https://github.com/BioMedIA/dHCP-release-notes/tree/master/supplementary_files).
84 |
85 | ## References
86 |
87 | 1. Bastiani, M., Andersson, J.L.R., Cordero-Grande, L., Murgasova, M.,
88 | Hutter, J., Price, A.N., Makropoulos, A., Fitzgibbon, S.P., Hughes,
89 | E., Rueckert, D., Victor, S., Rutherford, M., Edwards, A.D., Smith,
90 | S.M., Tournier, J.D., Hajnal, J.V., Jbabdi, S., and Sotiropoulos,
91 | S.N. **Automated processing pipeline for neonatal diffusion MRI in the
92 | developing Human Connectome Project** *Neuroimage (2019), 185: 750-763.* [DOI:
93 | 10.1016/j.neuroimage.2018.05.064](https://doi.org/10.1016/j.neuroimage.2018.05.064)
94 |
95 | 2. Andersson, J.L., Skare, S., and Ashburner, J. **How to correct
96 | susceptibility distortions in spin-echo echo-planar images: application
97 | to diffusion tensor imaging** *Neuroimage (2003), 20(2): 870-888.* [DOI:
98 | 10.1016/S1053-8119(03)00336-7](https://doi.org/10.1016/S1053-8119(03)00336-7)
99 |
100 | 3. Andersson, J.L., and Sotiropoulos, S.N. **An integrated approach
101 | to correction for off-resonance effects and subject movement in
102 | diffusion MR imaging** *Neuroimage (2016), 125: 1063-1078.* [DOI:
103 | 10.1016/j.neuroimage.2015.10.019](https://doi.org/10.1016/j.neuroimage.2015.10.019)
104 |
105 | 4. Andersson, J.L.R., Graham, M.S., Drobnjak, I., Zhang, H.,
106 | and Campbell, J. **Susceptibility-induced distortion that varies
107 | due to motion: Correction in diffusion MR without acquiring
108 | additional data** *NeuroImage (2018), 171: 277-295.* [DOI:
109 | 10.1016/j.neuroimage.2017.12.040](https://doi.org/10.1016/j.neuroimage.2017.12.040)
110 |
111 | 5. Andersson, J.L.R., Graham, M.S., Drobnjak, I., Zhang, H.,
112 | Filippini, N., and Bastiani, M. **Towards a comprehensive framework
113 | for movement and distortion correction of diffusion MR images:
114 | Within volume movement** *Neuroimage (2017), 152: 450-466.* [DOI:
115 | 10.1016/j.neuroimage.2017.02.085](https://doi.org/10.1016/j.neuroimage.2017.02.085)
116 |
117 | 6. Andersson, J.L.R., Graham, M.S., Zsoldos, E., and Sotiropoulos,
118 | S.N. **Incorporating outlier detection and replacement into a
119 | non-parametric framework for movement and distortion correction
120 | of diffusion MR images** *Neuroimage (2016), 141: 556-572.* [DOI:
121 | 10.1016/j.neuroimage.2016.06.058](https://doi.org/10.1016/j.neuroimage.2016.06.058)
122 |
123 | 7. Kuklisova-Murgasova, M., Quaghebeur, G., Rutherford, M.A., Hajnal, J.V.,
124 | and Schnabel, J.A. **Reconstruction of fetal brain MRI with intensity matching
125 | and complete outlier removal** *Med Image Anal (2012), 16(8): 1550-1564.* [DOI:
126 | 10.1016/j.media.2012.07.004](https://doi.org/10.1016/j.media.2012.07.004)
127 |
128 | 8. Greve, D.N., and Fischl, B. **Accurate and robust brain image alignment
129 | using boundary-based registration** *Neuroimage (2009), 48(1): 63-72.* [DOI:
130 | 10.1016/j.neuroimage.2009.06.060](https://doi.org/10.1016/j.neuroimage.2009.06.060)
131 |
132 | 9. Jenkinson, M., Bannister, P., Brady, M., and Smith, S. **Improved
133 | optimization for the robust and accurate linear registration and motion
134 | correction of brain image** *Neuroimage (2002), 17(2): 825-841.* [DOI:
135 | 10.1006/nimg.2002.1132](https://doi.org/10.1006/nimg.2002.1132)
136 |
137 | 10. Andersson, J.L.R., Jenkinson, M., and
138 | Smith, S. [**Non-linear registration, aka spatial
139 | normalisation**](https://www.fmrib.ox.ac.uk/datasets/techrep/tr07ja2/tr07ja2.pdf)
140 | *FMRIB technical report TR07JA2 (2010).*
141 |
142 | 11. Schuh, A., Makropoulos, A., Robinson, E.C., Cordero-Grande, L., Hughes,
143 | E., Hutter, J., Price, A.N., Murgasova, M., Teixeira, R.P.A.G., Tusor,
144 | N., Steinweg, J.K., Victor, S., Rutherford, M.A., Hajnal, J.V., Edwards,
145 | A.D., and Rueckert, D. **Unbiased construction of a temporally consistent
146 | morphological atlas of neonatal brain development** *bioRxiv (2018), 251512.*
147 | [DOI: 10.1101/251512](https://doi.org/10.1101/251512)
148 |
149 | 12. Bastiani, M., Cottaar, M., Fitzgibbon, S.P., Suri, S.,
150 | Alfaro-Almagro, F., Sotiropoulos, S.N., Jbabdi, S., and Andersson,
151 | J.L.R. **Automated quality control for within and between studies
152 | diffusion MRI data using a non-parametric framework for movement
153 | and distortion correction** *Neuroimage (2019), 184: 801-812.* [DOI:
154 | 10.1016/j.neuroimage.2018.09.073](https://doi.org/10.1016/j.neuroimage.2018.09.073)
155 |
156 | 13. Jbabdi S, Sotiropoulos S.N., Savio A., Grana M., Behrens
157 | T.E.J. **Model-based analysis of multishell diffusion MR data for
158 | tractography: How to get over fitting problems** *Magn Reson Med (2012).*
159 | [DOI:10.1002/mrm.24204](https://doi.org/10.1002/mrm.24204)
160 |
161 |
--------------------------------------------------------------------------------
/recon.md:
--------------------------------------------------------------------------------
1 | ---
2 | ---
3 |
4 | ## Reconstruction pipeline
5 |
6 | The reconstruction pipeline was developed by the team at King’s College
7 | London.
8 |
9 | Motion corrected volumetric reconstructions of multi-slice inversion
10 | recovery T1- and turbo spin echo T2-weighted images are obtained by
11 | extending the aligned sensitivity encoding (SENSE) method1
12 | to the multi-slice case2. Corrections are performed
13 | for both within-plane and through-plane motion from partial
14 | k-space information. The two acquired orthogonal stacks are
15 | integrated by a super-resolution scheme3. Methods and example
16 | data for the aligned SENSE method are available at
17 | https://github.com/mriphysics/multiSliceAlignedSENSE/releases/tag/1.0.1.
18 |
19 | ### Inputs and outputs
20 |
21 | **Path:** `rawdata/sub-{subid}/ses-{sesid}`
22 |
23 | The reconstruction pipeline generates the files listed in the [Reconstruction pipeline](structure.html#reconstruction-pipeline)
24 | section of the directory structure summary.
25 |
26 | Standard magnitude and phase reconstructions are provided for all native
27 | acquired anatomical image stacks, with the different orientations and repeat
28 | acquisitions labelled by sequence run number as follows:
29 |
30 | | Description | Filename |
31 | |:-----------------------------------------------------|:-----------------------------------------------------|
32 | | T1w magnitude image (native acquired stack) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_T1w.nii` |
33 | | T1w phase image (native acquired stack) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-phase_T1w.nii` |
34 | | T2w magnitude image (native acquired stack) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_T2w.nii` |
35 | | T2w phase image (native acquired stack) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-phase_T2w.nii` |
36 |
37 | Motion corrected and super-resolved reconstructions, and phase images, are included for every acquired multislice stack:
38 |
39 | | Description | Filename |
40 | |:---------------------------------|:------------------------------------------------|
41 | | T1w image (motion corrected) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-mc_T1w.nii` |
42 | | T2w image (motion corrected) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-mc_T2w.nii` |
43 | | T1w image (motion corrected and super resolved) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-mcsr_T1w.nii` |
44 | | T2w image (motion corrected and super resolved) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-mcsr_T2w.nii` |
45 |
46 | ### Primary anatomical outputs
47 | The final motion corrected slice-to-volume reconstructed T1w and T2w volumes are:
48 |
49 | | Description | Filename |
50 | |:--------------------------------------------------------|:------------------------------------------------|
51 | | T1w image (combined Slice-to-Volume reconstruction) | `anat/sub-{subid}_ses-{sesid}_rec-SVR_T1w.nii` |
52 | | T2w image (combined Slice-to-Volume reconstruction) | `anat/sub-{subid}_ses-{sesid}_rec-SVR_T2w.nii` |
53 |
54 | An additional T1 3D MPRAGE volume is also provided, but currently not used in processing pipelines:
55 |
56 | | Description | Filename |
57 | |:--------------------------------------------------------|:------------------------------------------------|
58 | | T1w image (3D MPRAGE) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_acq-MPRAGE_T1w.nii` |
59 |
60 | For fMRI and dMRI, simultaneous multi-slice (SMS) echo planar
61 | imaging (EPI) is reconstructed using the extended SENSE technique4,
62 | with details described elsewhere5,6,7; sensitivity estimates
63 | from a conventional reference scan are refined with the information from
64 | non-SMS reference acquisitions with matched readouts to promote matched
65 | coil map and image distortions. As for dMRI, complex data retrieval is
66 | performed by the generalized singular value shrinkage (GSVS) denoising
67 | technique using noise measures performed during the acquisition8.
68 | Methods and example data for the GSVS method are available at
69 | https://github.com/mriphysics/complexSVDShrinkageDWI/releases/tag/1.1.0.
70 |
71 |
72 | ### Primary dMRI outputs
73 | dMRI reconstructions are provided with and without denoising, along with phase images, and additional chi^2
74 | maps from the reconstruction required for the dMRI (SHARD) pipeline.
75 | A third reconstruction used by the dMRI (EDDY) pipeline is included, this matches the version from the 2nd data
76 | release, the main difference being that residual fat artefacts are suppressed in the reconstruction pipeline:
77 |
78 | | Description | Filename |
79 | |:--------------------------------------------------------|:------------------------------------------------|
80 | | Multi-band dMRI EPI | `dwi/sub-{subid}_ses-{sesid}_run-{seqnum}_dwi.nii` |
81 | | Multi-band dMRI EPI (denoised reconstruction) | `dwi/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-denoised_dwi.nii` |
82 | | Multi-band dMRI EPI (Release 2 reconstruction) | `dwi/sub-{subid}_ses-{sesid}_rec-release2_dwi.nii` |
83 |
84 | ### Primary outputs for fMRI
85 | There is one primary reconstruction for the resting state fMRI data, along with single-band reference scans (typically one before and
86 | one after resting state run), and a spin-echo EPI with matched readout for field estimation. In addition, all phase images are provided.
87 |
88 | | Description | Filename |
89 | |:--------------------------------------------------------|:------------------------------------------------|
90 | | Single-band Ref | `func/sub-{subid}_ses-{sesid}_run-{seqnum}_task-rest_sbref.nii` |
91 | | Resting fMRI | `func/sub-{subid}_ses-{sesid}_run-{seqnum}_task-rest_bold.nii` |
92 | | 4D Spin Echo EPI with different phase encode directions (for topup fieldmap estimation) | `fmap/sub-{subid}_ses-{sesid}_run-{seqnum}_epi.nii` |
93 |
94 | ### Primary outputs for field mapping
95 | Data from calibration scans are provided for B1+ and B0 field estimates,
96 | using the DREAM and dual-gradient echo methods, respectively. Reconstructed magnitude and phase
97 | images are provided, along with the following calculated field maps as the primary outputs:
98 |
99 | | Description | Filename |
100 | |:-----------------------------------------------------|:------------------------------------------------|
101 | | B0 field-map in (Hz) - unfiltered | `fmap/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-raw_fieldmap.nii` |
102 | | B1+ field map (rel. nom. flip) | `B1/sub-{subid}_ses-{sesid}_run-{seqnum}_b1map.nii` |
103 |
104 |
105 | The full output list can be found in the [Reconstruction pipeline](structure.html#reconstruction-pipeline)
106 | section of the directory structure summary.
107 |
108 |
109 | ## References
110 |
111 | 1. Cordero-Grande, L., Teixeira. R. P. A. G., Hughes, E. J.,
112 | Hutter, J., Price, A. N., and Hajnal, J. V. **Sensitivity encoding
113 | for aligned multishot magnetic resonance reconstruction** *IEEE
114 | Transactions on Computational Imaging (2016), 2(3): 266-280.* [DOI:
115 | 10.1109/TCI.2016.2557069](https://doi.org/10.1109/TCI.2016.2557069)
116 |
117 | 2. Cordero-Grande, L., Hughes, E. J., Hutter, J., Hutter, J., Price, A. N.,
118 | and Hajnal, J. V. **Three-Dimensional Motion Corrected Sensitivity Encoding
119 | Reconstruction for Multi-Shot Multi-Slice MRI: Application to Neonatal
120 | Brain Imaging** *Magnetic Resonance in Medicine 2018, 79(3): 1365-1376.*
121 | [DOI: 10.1002/mrm.26796](https://doi.org/10.1002/mrm.26796)
122 |
123 | 3. Kuklikova-Murgasova, M., Quaghebeur, G., Rutherford, M. A.,
124 | Hajnal, J. V., and Schnabel, J. A. **Reconstruction of fetal
125 | brain MRI with intensity matching and complete outlier removal**
126 | *Medical Image Analysis (2012), 16(8): 1550-1564.* [DOI:
127 | 10.1016/j.media.2012.07.004](https://doi.org/10.1016/j.media.2012.07.004)
128 |
129 | 4. Zhu, K., Dougherty, R. F., Wu, H., Middione, M. J., Takahashi,
130 | A. M, Zhang, T., Pauly, J. M., Kerr, A. B. **Hybrid-space
131 | SENSE reconstruction for simultaneous multi-slice MRI** *IEEE
132 | Transactions on Medical Imaging, 35(8) (2016):1824-1836.* [DOI:
133 | 10.1109/TMI.2016.2531635](https://doi.org/10.1109/TMI.2016.2531635)
134 |
135 | 5. Cordero-Grande, L., Price, A. N., and
136 | Hajnal, J. V. [**Comprehensive CG-SENSE reconstruction of
137 | SMS-EPI**](http://www.developingconnectome.org/wp-content/uploads/sites/70/2019/08/Comprehensive-CG-SENSE-reconstruction-of-SMS-EPI.pdf)
138 | *ISMRM 2016: 3239.*
139 |
140 | 6. Cordero-Grande, L., Hutter, J., Price,
141 | A., Hughes, E., and Hajnal, J. V. [**Goodness of
142 | fit factor in SENSE reconstruction: a tool for pseudolesion detection and fat
143 | unfolding**](http://www.developingconnectome.org/wp-content/uploads/sites/70/2019/08/Goodness-of-fit-factor-in-SENSE-reconstruction-a-tool-for-pseudolesion-detection-and-fat-unfolding.pdf)
144 | *ESMRMB 2016: 458.*
145 |
146 | 7. Hennel, F.,Buehrer, M., von Deuster, C., Seuven, A., and Pruessmann,
147 | K. P. **SENSE reconstruction for multiband EPI including slice-dependent N/2
148 | ghost correction** *Magnetic Resonance in Medicine (2016), 76(3): 873-879.*
149 | [DOI: 10.1002/mrm.25915](https://doi.org/10.1002/mrm.25915)
150 |
151 | 8. Cordero-Grande, L., Christiaens, D., Hutter, J., Price, A. N., and
152 | Hajnal, J. V. **Complex diffusion-weighted image estimation via matrix
153 | recovery under general noise models** *Neuroimage (2019), 200: 391-404.* [DOI:
154 | 10.1016/j.neuroimage.2019.06.039](https://doi.org/10.1016/j.neuroimage.2019.06.039)
155 |
156 |
157 |
--------------------------------------------------------------------------------
/dwi-shard.md:
--------------------------------------------------------------------------------
1 | ---
2 | ---
3 |
4 | ## The diffusion SHARD pipeline
5 |
6 | ### Inputs
7 |
8 | **From reconstruction pipeline:** `rawdata/sub-{subid}/ses-{sesid}`
9 |
10 | | Description | Filename |
11 | |:----------------------------------------------------------------------------------------|:--------------------------------------------------------------------------|
12 | | Multi-band dMRI EPI (denoised reconstruction) | `dwi/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-denoised_dwi.nii` |
13 | | Multi-band dMRI EPI (Chi2-maps) | `dwi/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-chi2_dwi.nii` |
14 |
15 |
16 | ### Outputs
17 |
18 | The primary outputs of the full SHARD preprocessing pipeline are:
19 |
20 | | Description | Filename |
21 | |------------------------------------------------------------------------------|-------------------------------------------------------------------------|
22 | | Preprocessed DWI data (denoising, motion correction, and destriping) | `dwi/sub-{subid}_ses-{sesid}_desc-preproc_dwi.nii.gz` |
23 | | List of b-values in FSL format | `dwi/sub-{subid}_ses-{sesid}_desc-preproc_dwi.bval` |
24 | | List of gradient directions in FSL format | `dwi/sub-{subid}_ses-{sesid}_desc-preproc_dwi.bvec` |
25 |
26 | (located in `derivatives/dhcp_dmri_shard_pipeline/sub-{subid}/ses-{sesid}`)
27 |
28 | The complete list of outputs and QC reports generated by the diffusion SHARD pipeline
29 | is listed in the [Diffusion SHARD pipeline](structure.html#diffusion-shard-pipeline)
30 | section of the directory structure summary.
31 |
32 |
33 | ### Pipeline
34 |
35 | The third data release includes an alternative dMRI processing
36 | pipeline based on denoising[1](#ref1) and SHARD motion
37 | correction[2](#ref2). This pipeline correponds more closely to
38 | the fetal processing pipeline[3](#ref3), and is therefore better
39 | suited for studies that include both the fetal and the neonatal data.
40 |
41 | The pipeline consists of the following processing steps:
42 |
43 | 1. Image denoising using random matrix theory with optimal
44 | shrinkage in the complex domain (i.e., using magnitude and phase
45 | images)[1](#ref1). This denoising approach also accounts
46 | for spatial noise correlations introduced by SENSE and Partial Fourier
47 | subsampling. After denoising, the magnitude images are extracted.
48 |
49 | 2. The denoised images are then corrected for Gibbs
50 | ringing[4](#ref4).
51 |
52 | 3. Fat shift suppression is achieved through local outlier reweighting in
53 | slice-to-volume reconstruction (step 5)[5](#ref5). The local
54 | outlier weights are computed independently based on the residuals of the
55 | SENSE reconstruction (Chi2-maps). This is achieved using a 2-class Gaussian
56 | Mixture Model with a Markov Random Field, fitted using Expectation-Maximization.
57 |
58 | 4. The B0 field map, used to model susceptibility-induced distortion,
59 | is estimated using FSL Topup[6](#ref6). As input for topup, we
60 | selected the two "best" b=0 volumes for each of the 4 phase encoding (PE)
61 | directions based on a edge-detection filter in the slice direction. The
62 | indices of the selected 8 volumes are stored and used as reference; the
63 | ordering of the image volumes and the corresponding diffusion encoding is
64 | thus never changed.
65 |
66 | 5. Motion correction using SHARD slice-to-volume
67 | reconstruction[2](#ref2). The inputs are the denoised and
68 | degibbsed multiband dMRI images (stage 2), the voxel weights (step 3),
69 | and the field map (step 4). The output are estimated subject motion traces,
70 | slice weights used to correct dropouts, and the SHARD representation fitted
71 | to the scattered slice data. The SHARD reconstruction also modelles the
72 | slice profile to recover the images at isotropic resolution. The motion-
73 | and distortion-corrected image is stored as a series of Spherical Harmonics
74 | (SH) coefficients for each shell.
75 |
76 | 6. The 5D SHARD output image is projected onto the original diffusion encoding
77 | used during acquisition, to provide the dMRI output in a format compatible with
78 | conventional software. This is a one-to-one mapping representing identical
79 | image information. Note that motion-induced gradient reorientation is modelled
80 | during SHARD slice-to-volume reconstruction in step 5; the gradient table
81 | hence remains unchanged.
82 |
83 | 7. Inter-slice intensity inhomogeneities were subsequently estimated and
84 | corrected on the motion-corrected projected dMRI volumes from step 6 using
85 | *dStripe*[7](#ref7). Note that these corrections were applied
86 | to the reconstructed dMRI data, not to the 5D SHARD output image.
87 |
88 | 8. Rigid alignement to high-resolution structural (T2-weighted) space
89 | using normalised mutual information (NMI) based registration with FSL
90 | Flirt [8](#ref8)on the mean b=1000 s/mm2 shell. This
91 | transformation is combined with a non-linear registration[9](#ref9)
92 | of the T2w volume to the 40 weeks template[10](#ref10) to allow
93 | transformations between diffusion and atlas spaces.
94 |
95 |
96 | ### Diffusion MRI QC
97 |
98 | Automated quality control metrics are calculated for several key steps in
99 | the pipeline. Specifically, the data release includes estimates of:
100 |
101 | - **Signal-to-Noise Ratio**: A measure of SNR is calculated based on the
102 | residuals of denoising (Step 1). The reported metrics are the median
103 | across a brain mask of the RMS residuals, divided by the median of the
104 | mean b=0 signal.
105 |
106 | - **Motion metrics**: Measures of the mean translation and rotation that was
107 | detected/estimated in the motion correction method.[2](#ref2)
108 | These metrics are based on the gradient of the motion trace, thus measuring
109 | the subject activity during the scan.
110 |
111 | - **Outlier ratio**: The mean outlier weight of all slices in the data, as
112 | detected in slice-to-volume reconstruction.[2](#ref2)
113 |
114 | In addition, overview screenshots of the motion traces and destriped output
115 | data are generated. Based on these screenshots, visual pass/fail Quality
116 | Control identified a small subset of cases to be discarded from analysis.
117 |
118 | All QC metrics are available in the `combined.tsv` spreadsheet in the
119 | [supplementary](https://github.com/BioMedIA/dHCP-release-notes/tree/master/supplementary_files).
120 |
121 |
122 | ### References
123 |
124 | 1. Cordero-Grande, L., Christiaens, D., Hutter, J., Price,
125 | A.N., Hajnal, J.V. **Complex diffusion-weighted image estimation via matrix
126 | recovery under general noise models** *Neuroimage (2019), 200: 391-404.* [DOI:
127 | 10.1016/j.neuroimage.2019.06.039](https://doi.org/10.1016/j.neuroimage.2019.06.039)
128 |
129 | 2. Christiaens, D., Cordero-Grande, L., Pietsch, M.,
130 | Hutter, J., Price, A.N., Hughes, E.J., Vecchiato, K., Deprez, M., Edwards,
131 | A.D., Hajnal, V., Tournier, J-D. **Scattered slice SHARD reconstruction
132 | for motion correction in multi-shell diffusion MRI** *NeuroImage (2021),
133 | 225: 117437.* [DOI:
134 | 10.1016/j.neuroimage.2020.117437](https://doi.org/10.1016/j.neuroimage.2020.117437)
135 |
136 | 3. Christiaens, D., Cordero-Grande, L., Price, A.N.,
137 | Hutter, J., Hughes, E.J., ounsell, S.J., Tournier, J-D., Hajnal, J.V. **Fetal
138 | diffusion MRI acquisition and analysis in the developing Human Connectome
139 | Project** *ISMRM 2020, O629.*
140 |
141 | 4. Kellner, E., Dhital, B., Kiselev, V.G., Reisert,
142 | M. **Gibbs-ringing artifact removal based on local subvoxel-shifts.**
143 | *Magnetic Resonance in Medicine (2016) 76: 1574–1581.* [DOI:
144 | 10.1002/mrm.26054](https://doi.org/10.1002/mrm.26054)
145 |
146 | 5. Christiaens, D., Cordero-Grande, L., Hutter, J.,
147 | Price, A.N., O'Murchearthaigh, J., Vecchiato, K., Hajnal, J.V., Tournier,
148 | J-D. **Fat-shift suppression in diffusion MRI using rotating phase encoding
149 | and localised outlier weighting** *ISMRM 2020, O981.*
150 |
151 | 6. Andersson, J.L., Skare, S., and
152 | Ashburner, J. **How to correct susceptibility distortions
153 | in spin-echo echo-planar images: application to diffusion
154 | tensor imaging** *Neuroimage (2003), 20(2): 870-888.* [DOI:
155 | 10.1016/S1053-8119(03)00336-7](https://doi.org/10.1016/S1053-8119(03)00336-7)
156 |
157 | 7. Pietsch, M. and Christiaens, D. and Hajnal, J.V. &
158 | Tournier, J-D. **dStripe: slice artefact correction in diffusion
159 | MRI via constrained neural network** *biorxiv (2020)* [DOI:
160 | 10.1101/2020.10.20.347518](https://doi.org/10.1101/2020.10.20.347518)
161 |
162 | 8. Jenkinson, M., Bannister, P., Brady, M., and Smith,
163 | S. **Improved optimization for the robust and accurate linear registration
164 | and motion correction of brain image** *Neuroimage (2002), 17(2): 825-841.*
165 | [DOI: 10.1006/nimg.2002.1132](https://doi.org/10.1006/nimg.2002.1132)
166 |
167 | 9. Andersson, J.L.R., Jenkinson,
168 | M., and Smith, S. [**Non-linear registration, aka spatial
169 | normalisation**](https://www.fmrib.ox.ac.uk/datasets/techrep/tr07ja2/tr07ja2.pdf)
170 | *FMRIB technical report TR07JA2 (2010).*
171 |
172 | 10. Schuh, A., Makropoulos, A., Robinson, E.C.,
173 | Cordero-Grande, L., Hughes, E., Hutter, J., Price, A.N., Murgasova,
174 | M., Teixeira, R.P.A.G., Tusor, N., Steinweg, J.K., Victor, S.,
175 | Rutherford, M.A., Hajnal, J.V., Edwards, A.D., and Rueckert,
176 | D. **Unbiased construction of a temporally consistent morphological
177 | atlas of neonatal brain development** *bioRxiv (2018), 251512.* [DOI:
178 | 10.1101/251512](https://doi.org/10.1101/251512)
179 |
180 |
181 | ### Pipeline scripts
182 |
183 | The processing scripts used in the dMRI SHARD pipeline are published on [GitHub](https://github.com/dchristiaens/dhcp-shard-pipeline).
184 |
--------------------------------------------------------------------------------
/struct.md:
--------------------------------------------------------------------------------
1 | ---
2 | ---
3 |
4 | ## The structural pipeline
5 |
6 | The [structural
7 | pipeline](https://github.com/BioMedIA/dhcp-structural-pipeline) was
8 | encapsulated as a Docker container image and run via the OpenMOLE10
9 | platform on a local cluster.
10 |
11 | ### Inputs and outputs
12 |
13 | **From reconstruction pipeline:** `rawdata/sub-{subid}/ses-{sesid}`
14 |
15 | | Description | Filename |
16 | |:--------------------------------------------------------|:------------------------------------------------|
17 | | T1w image (combined Slice-to-Volume reconstruction) | `anat/sub-{subid}_ses-{sesid}_rec-SVR_T1w.nii` |
18 | | T2w image (combined Slice-to-Volume reconstruction) | `anat/sub-{subid}_ses-{sesid}_rec-SVR_T2w.nii` |
19 |
20 | The structural pipline generates the files listed in the [Structural pipeline](structure.html#structural-pipeline) section of the directory structure summary.
21 |
22 | ### Operation
23 |
24 | 1. Registration
25 |
26 | 2. Segmentation
27 |
28 | 1. Structural scans are pre-processed by first running bias correction
29 | using the N4 algorithm1.
30 |
31 | 2. Scans are then brain extracted using BET2 from FSL.
32 |
33 | 3. Segmentation of the T2w volume is performed using the DRAW-EM
34 | algorithm3. DRAW-EM is an atlas-based segmentation technique
35 | that segments the volumes into 87 regions (see region names). Manually
36 | labelled atlases, annotated by an expert neuroanatomist4, are
37 | registered to the volume and their labels are fused to the subject
38 | space to provide structure priors. Segmentation is then performed with
39 | an Expectation-Maximization scheme that combines the structure priors
40 | and an intensity model of the volume. The 87 regions are further merged
41 | to provide the tissue segmentation (see tissue types).
42 |
43 | 4. All T1 weighted images have been pre-aligned to the T2w volumes
44 | using rigid alignment.
45 |
46 | 5. Both T1w and T2w volumes are defaced for anonymization based on
47 | registration and transformation of a manually annotated face mask.
48 |
49 | 3. Surface extraction
50 |
51 | 1. Surface mesh extraction is performed with the method described
52 | elsewhere5. A white matter mask enclosing the white surface is
53 | computed by merging the white matter and the subcortical structures with
54 | the exception of the brainstem and the cerebellum. Similarly, a pial mask
55 | is computed by merging the grey matter structure with the white matter
56 | mask. The white and pial surfaces of the left and right hemispheres
57 | are then reconstructed with the method outlined in 5 using a deformable
58 | model. The model in 5 includes forces to avoid self-intersections and
59 | includes an image-based refinement step that corrects regions such as
60 | deep sulci mislabelled by the volumetric segmentation.
61 |
62 | 2. Midthickness surfaces are generated as the middle surface between
63 | the white and pial surfaces. The midthickness surface is computed using
64 | the Euclidean distance between corresponding points of the white and
65 | pial surface.
66 |
67 | 3. Spherical projection is performed6, and it is based on the inflated
68 | white matter surface. The inflated white matter surface is produced in
69 | a similar manner as in the FreeSurfer pipeline7. Inflated and
70 | very inflated surfaces used for visualisation are generated
71 | similarly8.
72 |
73 | 4. The following metrics are further estimated from the surfaces:
74 | curvature, thickness, sulcal depth, T1w/T2w myelin, labels (projected
75 | from the volume). All surfaces have one-to-one vertex correspondence
76 | for all points on the surface ensuring that the same vertex indexes
77 | the same point, in the same relative position, on the anatomy for all
78 | surfaces. Please note that due to the relatively large voxels (causing
79 | partial volume) and/or the uneven vertex sampling for gyri (relative
80 | to sulci) we observe some artifacts in the myelin and thickness metric
81 | files, which resemble the folding patterns. For this, we offer a
82 | 'corrected' version of thickness maps (corr_thickness); however, this
83 | results only from a linear regressesion based correction. We advise
84 | careful interpretation of these maps and have chosen at this time not
85 | to do the same for myelin.
86 |
87 | 4. Surface registration
88 |
89 | 1. A new symmetric and extended version of the neonatal surface
90 | template11 is available from the [brain-development.org
91 | website](https://brain-development.org/brain-atlases/atlases-from-the-dhcp-project/)
92 |
93 | 2. All surfaces have been nonlinearly aligned to the template using
94 | cortical folding-driven aligment (implemented with MSM11,12);
95 | alignment has been optimised with relatively weak regularisation
96 | to push cortical correspondence of folds in the frontal lobe. At
97 | this time we have no evidence of any negative impact on the dMRI
98 | and fMRI correspondence through doing this. However, scripts to
99 | re-run registration wih modifed parameters are available from
100 | [here](https://github.com/ecr05/dHCP_template_alignment)
101 |
102 | 3. We release only the registration warp file (`sphere.reg.surf.gii`). To
103 | obtain the warped metric files and surfaces in template space please run
104 | the [resampling scripts](https://github.com/ecr05/dHCP_template_alignment)
105 |
106 |
107 | The structural pipeline is described in detail elsewhere9.
108 |
109 |
110 | ### Quality Control/Assurance
111 |
112 | The dHCP structural data set contains MR imaging of whole brain anatomy across a wide gestational age. The data collected contains rapidly changing anatomical variation across age, that inherently provides challenges for automatic segmentation processes. For this reason, we carried out a visual inspection of the segmentation pipeline on a random set of 30 datasets (gestational age range at scan 27.14 weeks-44.14 weeks).
113 |
114 | The following anatomical segmentations were assessed:
115 | - Cortical ribbon
116 | - Ventricular space including the cavum septum pellucidum
117 | - Whole brain white matter
118 | - Deep grey matter (basal ganglia)
119 |
120 | The results showed some small areas of common segmentation errors across the dataset that can be easily correctable with some fine editing:
121 | 1. The cortical ribbon segmentation can contain small regions of highly convoluted cortex that may not be sufficiently delineated due to partial volume effects and there is consistent mis-registration of the cortex in the medial temporal lobe.
122 | 2. The anterior cavum septum pellucidum can be mis-labelled as ventricular space.
123 | 3. The basal ganglia/deep grey matter anatomy may have a small chunk of anatomy mislabelled as white matter.
124 | 4. The whole brain white matter segmentation was found to be consistently accurate across subjects.
125 |
126 | We therefore recommend that visual inspection of the segmentation data in this cohort is carried out before inclusion in analysis.
127 |
128 |
129 |
130 | ### References
131 |
132 | 1. Tustison, N. J., Avants, B. B., Cook, P. A., Zheng, Y., Egan, A.,
133 | Yushkevich, P. A. and Gee, J. C. **N4ITK: improved N3 bias correction. IEEE
134 | transactions on medical imaging (2010)** *29(6): 1310-1320.* [DOI:
135 | 10.1109/TMI.2010.2046908](https://doi.org/10.1109/TMI.2010.2046908)
136 |
137 | 2. Smith, S. M. **Fast robust automated brain extraction**
138 | *Human Brain Mapping (2002), 17(3): 143-55.* [DOI:
139 | 10.1002/hbm.10062](https://doi.org/10.1002/hbm.10062)
140 |
141 | 3. Makropoulos, A., Gousias I. S., Ledig C., Aljabar P., Serag A,
142 | Hajnal J. V., Edwards A. D., Counsell S. J., and Rueckert D. **Automatic
143 | whole brain MRI segmentation of the developing neonatal brain** *IEEE
144 | transactions on medical imaging (2014), 33(9): 1818-1831.* [DOI:
145 | 10.1109/TMI.2014.2322280](https://doi.org/10.1109/TMI.2014.2322280)
146 |
147 | 4. Gousias, I. S., Edwards A. D., Rutherford M. A., Counsell S. J., Hajnal
148 | J. V., Rueckert D., and Hammers, A. **Magnetic resonance imaging of the
149 | newborn brain: manual segmentation of labelled atlases in term-born
150 | and preterm infants** *Neuroimage (2012), 62(3): 1499-1509.* [DOI:
151 | 10.1016/j.neuroimage.2012.05.083](https://doi.org/10.1016/j.neuroimage.2012.05.083)
152 |
153 | 5. Schuh, A., Makropoulos, A., Wright, R., Robinson, E. C., Tusor, N.,
154 | Steinweg, J., Hughes, E., Cordero Grande, L., Price, A., Hutter, J., Hajnal,
155 | J., and Rueckert, D. **A deformable model for reconstruction of the neonatal
156 | cortex** *IEEE 14th International Symposium on Biomedical Imaging 2017.*
157 | [DOI: 10.1109/ISBI.2017.7950639](https://doi.org/10.1109/ISBI.2017.7950639)
158 |
159 | 6. Elad, A., Keller, Y., and Kimmel,
160 | R. [**Texture Mapping via Spherical Multidimensional
161 | Scaling**](http://www.developingconnectome.org/wp-content/uploads/sites/70/2019/08/Texture-Mapping-via-Spherical-Multidimensional-Scaling-.pdf)
162 | *International Conference on Scale-Space Theories in Computer Vision
163 | (2005): 443–455.*
164 |
165 | 7. Fischl, B., Sereno M. I., and Dale, A. M. **Cortical surface-based
166 | analysis: II: inflation, flattening, and a surface-based
167 | coordinate system** N*euroimage (1999), 9(2): 195-207.* [DOI:
168 | 10.1006/nimg.1998.0396](https://doi.org/10.1006/nimg.1998.0396)
169 |
170 | 8. Glasser, M., Sotiropoulo, S. N., Wilson, J. A, Coalson, T. S,
171 | Fischl, B., Andersson, L., Xu, J., Jbabdi, S., Webster, M.,
172 | Polimeni, J. R., Van Essen, D. C., Jenkinson, M., WU-Minn HCP
173 | Consortium. **The minimal preprocessing pipelines for the Human
174 | Connectome Project** *NeuroImage (2013)., 80: 105–124.* [DOI:
175 | 10.1016/j.neuroimage.2013.04.127](https://doi.org/10.1016/j.neuroimage.2013.04.127)
176 |
177 | 9. Makropoulos, A., Robinson, E.C., Schuh, A., Wright, R., Fitzgibbon,
178 | S., Bozek, J., Counsell, S. J., Steinweg, J., Vecchiato, K.,
179 | Passerat-Palmbach, J., Lenz, G., Mortari, F., Tenev, T., Duff, E. P.,
180 | Bastiani, M., Cordero-Grande, L., Hughes, E., Tusor, N., Tournier,
181 | J. D., Hutter, J., Price, A. N., Teixeira, R. P. A. G., Murgasova M,
182 | Victor, S., Kelly, C., Rutherford, M. A., Smith, S. M., Edwards, A. D.,
183 | Hajnal, J. V., Jenkinson, M., and Rueckert, D. **The developing Human
184 | Connectome Project: a Minimal Processing Pipeline for Neonatal Cortical
185 | Surface Reconstruction** *NeuroImage (2018), 173: 88-112.* [DOI:
186 | 10.1016/j.neuroimage.2018.01.054](https://doi.org/10.1016/j.neuroimage.2018.01.054)
187 |
188 | 10. Passerat-Palmbach, J., Reuillon, R., Leclaire, M., Makropoulos,
189 | A., Robinson, E.C., Parisot, S., and Rueckert, D. **Reproducible
190 | Large-Scale Neuroimaging Studies with the OpenMOLE Workflow
191 | Management System** *Frontiers in Neuroinformatics (2017), 11.* [DOI:
192 | 10.3389/fninf.2017.00021](https://doi.org/10.3389/fninf.2017.00021)
193 |
194 | 11. Bozek, J., Makropoulos, A., Schuh, A., Fitzgibbon, S., Wright, R.,
195 | Glasser, M.F., Coalson, T.S., O'Muircheartaigh, J., Hutter, J., Price,
196 | A.N., Cordero-Grande, L. Teixeira, R. P. A. G., Hughes, E., Tusor, N.,
197 | Pegoretti Baruteau, K., Rutherford, M. A., Edwards, A. D., Hajnal, J. V.,
198 | Smith, S. M., Rueckert, D., Jenkinson, M., Robinson, E.C, **Construction of
199 | a neonatal cortical surface atlas using Multimodal Surface Matching in the
200 | Developing Human Connectome Project**. *NeuroImage (2018) 179: 11-29.* [DOI:
201 | 10.1016/j.neuroimage.2018.06.018](https://doi.org/10.1016/j.neuroimage.2018.06.018)
202 |
203 | 12. Robinson, E.C., Jbabdi, S., Glasser, M.F., Andersson, J.,
204 | Burgess, G.C., Harms, M.P., Smith, S.M., Van Essen, D.C. and
205 | Jenkinson, M. **MSM: a new flexible framework for Multimodal
206 | Surface Matching**. *Neuroimage (2014), 100: 414-26.* [DOI:
207 | 10.1016/j.neuroimage.2014.05.069](https://doi.org/10.1016/j.neuroimage.2014.05.069)
208 |
209 | 13. Robinson, E.C., Garcia, K., Glasser, M.F., Chen, Z., Coalson, T.S.,
210 | Makropoulos, A., Bozek, J., Wright, R., Schuh, A., Webster, M. and Hutter,
211 | J., Price, A.N., Cordero-Grande, L. Hughes, E., Tusor, N., Bayley, P.V.,
212 | Van Essen, D.C., Smith, S. M., Edwards, A. D., Hajnal, J. Jenkinson, M.,
213 | Glocker, B., Rueckert, D., **Multimodal surface matching with higher-order
214 | smoothness constraints**. *Neuroimage (2018), 167: 453-465.* [DOI:
215 | 10.1016/j.neuroimage.2017.10.037](https://doi.org/10.1016/j.neuroimage.2017.10.037)
216 |
217 |
--------------------------------------------------------------------------------
/fmri.md:
--------------------------------------------------------------------------------
1 | ---
2 | ---
3 |
4 | ## Functional (fMRI) Pipeline
5 |
6 | ### Inputs
7 |
8 | **From reconstruction pipeline:** `rawdata/sub-{subid}/ses-{sesid}`
9 |
10 | | Description | Filename |
11 | |:----------------------------------------------------------------------------------------|:--------------------------------------------------------------------------|
12 | | Resting fMRI | `func/sub-{subid}_ses-{sesid}_run-{seqnum}_task-rest_bold.nii` |
13 | | Single-band Ref | `func/sub-{subid}_ses-{sesid}_run-{seqnum}_task-rest_sbref.nii` |
14 | | 4D Spin Echo EPI with different phase encode directions (for topup fieldmap estimation) | `fmap/sub-{subid}_ses-{sesid}_run-{seqnum}_epi.nii` |
15 | | Dual echo-time (magnitude) | `fmap/sub-{subid}_ses-{sesid}_run-{seqnum}_magnitude.nii` |
16 | | Dual echo-time field-map in (Hz) - filtered and smoothed | `fmap/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-filtered_fieldmap.nii` |
17 |
18 | **From structural pipeline:** `derivatives/dhcp_anat_pipeline/sub-{subid}/ses-{sesid}`
19 |
20 | | Description | Filename |
21 | |:-------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------|
22 | | T2 weighted, bias corrected image | `anat/sub-{subid}_ses-{sesid}_desc-restore_T2w.nii.gz` |
23 | | Draw-EM tissue segmentation (9 labels) | `anat/sub-{subid}_ses-{sesid}_desc-drawem9_dseg.nii.gz` |
24 | | Left/Right white surface | `anat/sub-{subid}_ses-{sesid}_hemi-{hemi}_wm.surf.gii` |
25 | | Left/Right pial surface | `anat/sub-{subid}_ses-{sesid}_hemi-{hemi}_pial.surf.gii` |
26 |
27 |
28 | ### Outputs
29 |
30 | The primary outputs of the fMRI preprocessing pipeline are:
31 |
32 | **Path:** `derivatives/dhcp_fmri_pipeline/sub-{subid}/ses-{sesid}`
33 |
34 | | Description | Filename |
35 | |:-------------------------------------------------------------------------------|:------------------------------------------------------------------------------|
36 | | Multi-band EPI, distortion corrected, motion corrected, FIX denoised, 4D image | `func/sub-{subid}_ses-{sesid}_task-rest_desc-preproc_bold.nii.gz` |
37 | | Warp from functional space to the extended dHCP 40-week template space | `xfm/sub-{subid}_ses-{sesid}_from-bold_to-extdhcp40wk_mode-image.nii.gz` |
38 |
39 | The complete list of outputs and QC reports generated by the fMRI pipeline
40 | are listed in the [Functional pipeline](structure.html#functional-pipeline) section of the directory structure summary.
41 |
42 | ### Pipeline
43 |
44 | Version `1.2.0` of the dHCP neonatal fMRI pipeline was used for release 3 and can be downloaded here: https://git.fmrib.ox.ac.uk/seanf/dhcp-neonatal-fmri-pipeline/-/tree/1.2.0
45 |
46 | The preprocessing steps implemented in the pipeline are as follows:
47 |
48 | 1. Prepare fieldmaps for correction of susceptibility distortions
49 |
50 | 1. Estimate field map from the two “best” spin-echo volumes (1 per
51 | phase-encode direction) using FSL TOPUP1. “Best” is
52 | defined by smoothness in the z-dimension (stdev of the slice-to-slice
53 | difference in the z-dimension)
54 |
55 | 2. If visual inspection indicates that the spin-echo has significant
56 | motion contamination then use the dual-echo-time-derived fieldmap instead
57 | of the spin-echo-derived fieldmap
58 |
59 | 2. Registration
60 |
61 | 1. Boundary-based registration (BBR, FSL FLIRT3) of the
62 | fieldmap magnitude to the T2w structural
63 |
64 | 2. Boundary-based registration (BBR, FSL FLIRT3) of the
65 | sbref to the T2 structural incorporating field map-based distortion
66 | correction of the sbref
67 |
68 | 3. Linear registration (6-dof, corratio, FSL FLIRT3) of the
69 | first volume of the functional multiband EPI to the sbref
70 |
71 | 4. After susceptibility and motion correction (see below), linear
72 | registration (6-dof, corratio, FSL FLIRT3) of the temporal
73 | mean of the motion and distortion corrected functional multiband EPI
74 | to the distortion corrected sbref
75 |
76 | 5. Nonlinear diffeomorphic multimodal registration of the
77 | age-matched T2w and GM/WM tissue probability from the dHCP
78 | volumetric atlas7 to the subjects T2w and GM/WM tissue
79 | probability using ANTs SyN6. The GM/WM probability for
80 | each subject was calculated as the partial volume estimate from a tool
81 | called [Toblerone](https://toblerone.readthedocs.io/en/latest/). We have
82 | [augmented](https://git.fmrib.ox.ac.uk/seanf/dhcp-resources/-/blob/master/docs/dhcp-augmented-volumetric-atlas-extended.md)
83 | the dHCP volumetric atlas with week-to-week nonlinear transforms estimated
84 | using a diffeomorphic T2w registration (ANTs SyN). The appropriate
85 | transforms are then combined to yield a (40 week) template-to-structural
86 | transform
87 |
88 | 6. From these primary registrations the following composite transforms
89 | are calculated:
90 |
91 | 1. Fieldmap to native functional
92 |
93 | 2. Motion and distortion corrected functional to 40-week template
94 | from the dHCP volumetric atlas
95 |
96 | 3. Susceptibility and motion correction
97 |
98 | 1. Slice-to-volume motion correction and motion-by-susceptibility
99 | correction is performed using FSL EDDY2
100 |
101 | 4. ICA Denoising
102 |
103 | 1. Temporal high-pass filter (150s high-pass cutoff) and ICA denoising
104 | using FSL FIX4, pre-trained with manually-labelled data from
105 | 50 dHCP neonatal subjects, to identify artefactual ICs (accuracy: median
106 | TPR=98%, median TNR=96%). The ICA dimensionality was capped at 600 ICs.
107 |
108 | 2. Noise ICs and motion parameters regressed from motion and distortion
109 | corrected functional multiband EPI.
110 |
111 | > **Schematic of the dHCP fMRI neonatal pre-processing pipeline:** The
112 | > schematic is segregated into the 4 main conceptual processing stages by
113 | > coloured background; fieldmap pre-processing (red), susceptibility and motion
114 | > correction (orange), registration (green), and denoising (purple). Inputs
115 | > to the pipeline are grouped in the top row, and the main pipeline outputs
116 | > are grouped in the lower right. Blue filled rectangles with rounded corners
117 | > indicate processing steps, whilst black rectangles (with no fill) represent
118 | > data. The critical path is denoted by magenta connector arrows. (dc) =
119 | > distortion corrected; (mcdc) = motion and distortion corrected.
120 |
121 | [](assets/images/pipeline_schematic.png)
122 |
123 |
124 | ### Quality Control/Assurance
125 |
126 | 1. Numerous quality assurance metrics are calculated during the
127 | pre-processing. Six of these are specifically compared against the population
128 | distribution to flag outliers for manual inspection and potential exclusion:
129 |
130 | 1. Mean DVARS5 of the ICA denoised functional EPI; provides
131 | a snapshot of output signal quality
132 |
133 | 2. Mean tSNR of the ICA denoised functional EPI; provides a snapshot
134 | of output signal quality
135 |
136 | 3. Normalised mutual information of the source (moving) image, re-sampled
137 | to reference space, and the reference (fixed) image, for each of the
138 | primary registrations:
139 |
140 | 1. Fieldmap to structural T2w
141 |
142 | 2. Native functional to sbref
143 |
144 | 3. Motion and distortion corrected functional to sbref
145 |
146 | 4. Sbref to structural T2w
147 |
148 | 5. Age-matched atlas template T2w to native structural T2w
149 |
150 | 2. All QA measures were converted to Z-scores and flipped as necessary so
151 | that positive z-scores are good and negative bad. Subject/sessions with
152 | a z-score < -2.5 on any QC metric were flagged for further inspection.
153 |
154 | 3. All QC metrics are available in the `combined.tsv` spreadsheet in the
155 | [supplementary](https://github.com/BioMedIA/dHCP-release-notes/tree/master/supplementary_files).
156 |
157 | > **QC Scores:** 28 subject/sessions were flagged for futher inspection
158 | > (see figure). All flagged subject/sessions are included in the release.
159 |
160 | [](assets/images/fmri_qc_z_distns.png)
162 |
163 | ### How to cite
164 |
165 | Detailed instructions on how to cite can be found here:
166 | [http://www.developingconnectome.org/how-to-cite/](http://www.developingconnectome.org/how-to-cite/)
167 |
168 | **Primary citation for the fMRI pipeline:**
169 |
170 | Fitzgibbon, SP, Harrison, SJ, Jenkinson, M, Baxter, L, Robinson, EC,
171 | Bastiani, M, Bozek, J, Karolis, V, Cordero Grande, L, Price, AN, Hughes,
172 | E, Makropoulos, A, Passerat-Palmbach, J, Schuh, A, Gao, J, Farahibozorg,
173 | S, O'Muircheartaigh, J, Ciarrusta, J, O'Keeffe, C, Brandon, J, Arichi, T,
174 | Rueckert, D, Hajnal, JV, Edwards, AD, Smith, SM, \*Duff, E, \*Andersson, J
175 | **The developing Human Connectome Project automated functional processing
176 | framework for neonates.**, *NeuroImage (2020), 223: 117303*, 2020. [DOI:
177 | 10.1016/j.neuroimage.2020.117303](https://doi.org/10.1016/j.neuroimage.2020.117303)
178 | _\*Authors contributed equally._
179 |
180 | ```
181 | @article {Fitzgibbon766030,
182 | author = {Fitzgibbon, Sean P. and Harrison, Samuel J. and Jenkinson, Mark and Baxter, Luke and Robinson, Emma C. and Bastiani, Matteo and Bozek, Jelena and Karolis, Vyacheslav and Grande, Lucilio Cordero and Price, Anthony N. and Hughes, Emer and Makropoulos, Antonios and Passerat-Palmbach, Jonathan and Schuh, Andreas and Gao, Jianliang and Farahibozorg, Seyedeh-Rezvan and O{\textquoteright}Muircheartaigh, Jonathan and Ciarrusta, Judit and O{\textquoteright}Keeffe, Camilla and Brandon, Jakki and Arichi, Tomoki and Rueckert, Daniel and Hajnal, Joseph V. and Edwards, A. David and Smith, Stephen M. and Duff, Eugene and Andersson, Jesper},
183 | title = {The developing Human Connectome Project (dHCP) automated resting-state functional processing framework for newborn infants},
184 | elocation-id = {117303},
185 | year = {2020},
186 | doi = {10.1016/j.neuroimage.2020.117303},
187 | publisher = {Elsevier},
188 | URL = {https://doi.org/10.1016/j.neuroimage.2020.117303},
189 | eprint = {https://www.sciencedirect.com/science/article/pii/S1053811920307898/pdfft?md5=18806cf190a26f783de4bef456fe28b6&pid=1-s2.0-S1053811920307898-main.pdf},
190 | journal = {NeuroImage}
191 | }
192 | ```
193 |
194 | ### References
195 |
196 | 1. Andersson, J. L., Skare, S. and Ashburner, J. **How to correct
197 | susceptibility distortions in spin-echo echo-planar images: application
198 | to diffusion tensor imaging** *Neuroimage (2003), 20: 870-888.* [DOI:
199 | 10.1016/S1053-8119(03)00336-7](https://doi.org/10.1016/S1053-8119(03)00336-7)
200 |
201 | 2. Andersson, J. L. and Sotiropoulos, S. N. **An integrated approach
202 | to correction for off-resonance effects and subject movement in
203 | diffusion MR imaging** *Neuroimage (2016), 125: 1063-1078.* [DOI:
204 | 10.1016/j.neuroimage.2015.10.019](https://doi.org/10.1016/j.neuroimage.2015.10.019)
205 |
206 | 3. Jenkinson, M., and Smith, S. **A global optimisation
207 | method for robust affine registration of brain images**
208 | *Medical image analysis (2001), 5(2): 143–156.* [DOI:
209 | 10.1016/S1361-8415(01)00036-6](https://doi.org/10.1016/S1361-8415(01)00036-6)
210 |
211 | 4. Salimi-Khorshidi, G., Douad, G, Beckman, C. F., Glasser, M. F.,
212 | Griffanti, L., and Smith, S. M. **Automatic denoising of functional
213 | MRI data: combining independent component analysis and hierarchical
214 | fusion of classifiers** *NeuroImage (2014), 90: 449–468.* [DOI:
215 | 10.1016/j.neuroimage.2013.11.046](https://doi.org/10.1016/j.neuroimage.2013.11.046)
216 |
217 | 5. Power, J. D., Barnes, K. A., Snyder, A. Z., Schlaggar,
218 | B. L., and Petersen, S. E. **Spurious but Systematic
219 | Correlations in Functional Connectivity MRI Networks Arise from
220 | Subject Motion** *NeuroImage (2012), 59(3): 2142–54.* [DOI:
221 | 10.1016/j.neuroimage.2011.10.018](https://doi.org/10.1016/j.neuroimage.2011.10.018)
222 |
223 | 6. Avants, B.B., Epstein, C.L., Grossman, M., Gee, J.C. **Symmetric
224 | diffeomorphic image registration with cross-correlation:
225 | Evaluating automated labeling of elderly and neurodegenerative
226 | brain.** *Med. Image Anal. (2008), 12, 26–41.* [DOI:
227 | 10.1016/j.media.2007.06.004](https://doi.org/10.1016/j.media.2007.06.004)
228 |
229 | 7. Andreas Schuh, Antonios Makropoulos, Emma C. Robinson, Lucilio
230 | Cordero-Grande, Emer Hughes, Jana Hutter, Anthony N. Price, Maria
231 | Murgasova, Rui Pedro A. G. Teixeira, Nora Tusor, Johannes Steinweg,
232 | Suresh Victor, Mary A. Rutherford, Joseph V. Hajnal, A. David Edwards,
233 | and Daniel Rueckert. **Unbiased construction of a temporally consistent
234 | morphological atlas of neonatal brain development**, *bioRxiv, 2018.* [DOI:
235 | 10.1101/251512](https://doi.org/10.1101/251512)
236 |
237 |
--------------------------------------------------------------------------------
/cite.md:
--------------------------------------------------------------------------------
1 | ---
2 | ---
3 |
4 | ## dHCP Citations
5 |
6 | *How to acknowledge dHCP and cite dHCP publications if you have used data
7 | provided by the KCL-Imperial-Oxford developing HCP consortium.*
8 |
9 | As stipulated in the User Terms, authors of publications or presentations
10 | that use KCL-Imperial-Oxford developing Human Connectome Project (dHCP)
11 | data should acknowledge the funding sources and cite relevant publications
12 | that describe key methods used by the dHCP to acquire and process the data.
13 | This page provides guidance on both fronts.
14 |
15 |
16 |
17 | ## Acknowledge the Funding Source
18 |
19 | Papers, book chapters, books, posters, oral presentations, and all other
20 | printed and digital presentations of results derived from dHCP data should
21 | contain the following wording in the acknowledgments section:
22 |
23 | > Data were provided by the developing Human Connectome Project,
24 | > KCL-Imperial-Oxford Consortium funded by the European Research Council
25 | > under the European Union Seventh Framework Programme (FP/2007-2013) / ERC
26 | > Grant Agreement no. [319456]. We are grateful to the families who generously
27 | > supported this trial.
28 |
29 | ## Cite Relevant Publications
30 |
31 | The specific publications that are appropriate to cite will depend on what
32 | dHCP data you used in your study and the purposes for which you used the data.
33 | Here is an annotated list of publications that can guide your choices.
34 | They are grouped into categories and subcategories that relate to different
35 | aspects of data acquisition, pre-processing, and analysis. As additional
36 | publications become available, this list will be updated to include those
37 | that it may be relevant to cite.
38 |
39 | ### Publications relevant to dHCP primary datasets
40 |
41 | The publications in this section describe dHCP data acquisition methods
42 | that have been used to generate both the ‘raw’ NIFTI format and the
43 | pre-processed datasets that are available for download.
44 |
45 | **Overview Publication:**
46 |
47 | Hughes, E. J., Winchman, T., Padormo, F., Teixeira, R., Wurie, J., Sharma,
48 | M., Fox, M., Hutter, J., Cordero‐Grande, L., Price, A. N., Allsop, J.,
49 | Bueno‐Conde, J., Tusor, N., Arichi, T., Edwards, A. D., Rutherford,
50 | M. A., Counsell, S. J., and Hajnal, J. V. **A dedicated neonatal brain
51 | imaging system** *Magnetic Resonance Medicine (2017), 78(2): 794–804* [DOI:
52 | 10.1002/mrm.26462](https://doi.org/10.1002/mrm.26462)
53 |
54 | **Diffusion MRI data acquisition:** All dHCP diffusion data was acquired
55 | with a purpose designed multiband EPI acquisition featuring optimised
56 | multi-shell diffusion sensitisation scheme, gradient demand optimisation,
57 | restart capability with adjustable time setback to provide overlapping data,
58 | and use of all 4 phase encode directions.
59 |
60 | Hutter, J., Tournier, J.D., Price, A.N., Cordero-Grande, L., Hughes, E.J.,
61 | Bastiani, M., Sotiropoulos, S.N., Jbabdi, S., Andersson, J., Edwards, A.D.,
62 | & Hajnal, J.V. **Time-efficient and flexible design of optimised multi-shell
63 | HARDI diffusion** *Magnetic Resonance in Medicine (2018), 79 (3): 1276-1292.*
64 | [DOI: 10.1002/mrm.26765](https://doi.org/10.1002/mrm.26765)
65 |
66 | Tournier, J. D., Christiaens, D., Hutter, J., Price, A. N., Cordero-Grande,
67 | L., Hughes, E., Bastiani, M., Sotiropoulos, S. N., Smith S. M., Rueckert,
68 | D., Counsell, S. J., Edwards, A. D., Hajnal, J. V. **A data‐driven
69 | approach to optimising the encoding for multi‐shell diffusion MRI with
70 | application to neonatal imaging.** *NMR in Biomedicine. 2020; 33:e4348.*
71 | [doi: 10.1002/nbm.4348](https://doi.org/10.1002/nbm.4348)
72 |
73 | **Resting state functional MRI data acquisition:** All dHCP functional
74 | imaging acquisitions were obtained using an optimised multiband sequence
75 | tuned for neonatal, in particular by deploying a high multiband factor to
76 | achieve a repeat time short enough to avoid aliasing cardiac fluctuations
77 | into the fMRI signal. Phase optimised multiband pulses were used throughout.
78 |
79 | Price A. N., Cordero-Grande L., Malik S. J., Abaei M., Arichi T.,
80 | Hughes E. J., Rueckert D., Edwards A. D., Hajnal J. V. **Accelerated
81 | Neonatal fMRI Using Multiband EPI** [*In Proc ISMRM 2015:
82 | p3911.*](http://www.developingconnectome.org/wp-content/uploads/sites/70/2019/08/Accelerated-Neonatal-fMRI-using-Multiband-EPI.-ISMRM-2015.pdf)
83 |
84 | Malik, S. J., Price A. N., and Hajnal J. V. **Optimized Amplitude
85 | Modulated Multi-Band RF pulses** [*In Proc ISMRM 2015:
86 | p2398*](http://www.developingconnectome.org/wp-content/uploads/sites/70/2019/08/Optimized-Amplitude-Modulated-Muli-Band-RF-pulses.pdf)
87 |
88 | **Anatomical MRI - Motion corrected reconstruction:** All anatomical images
89 | for all dHCP subjects have had motion corrected reconstruction:
90 |
91 | Cordero-Grande, L., Hughes, E. J., Hutter, J., Hutter, J., Price, A. N.,
92 | and Hajnal, J. V. **Three-Dimensional Motion Corrected Sensitivity Encoding
93 | Reconstruction for Multi-Shot Multi-Slice MRI: Application to Neonatal
94 | Brain Imaging** *Magnetic Resonance in Medicine (2018), 79(3): 1365–1376.* [DOI:
95 | 10.1002/mrm.26796](https://doi.org/10.1002/mrm.26796)
96 |
97 | **Diffusion MRI signal retrieval:** Denoised diffusion images are obtained by:
98 |
99 | Cordero-Grande, L., Christiaens, D., Hutter, J., Price, A. N., and Hajnal,
100 | J. V. **Complex diffusion-weighted image estimation via matrix recovery
101 | under general noise models** *Neuroimage (2019), 200: 391–404.* [DOI:
102 | 10.1016/j.neuroimage.2019.06.039](https://doi.org/10.1016/j.neuroimage.2019.06.039)
103 |
104 | ### Publications relevant to the dHCP pre-processed data
105 |
106 | The publications in this section describe methods that are relevant if you
107 | have downloaded and used any of the dHCP pre-processed data involving one
108 | or more modalities.
109 |
110 | #### Structural MRI data processing
111 |
112 | **Automated processing pipeline:** All dHCP subjects have been processed and
113 | cortical meshes have been generated using the following automated pipeline:
114 |
115 | Makropoulos, A., Robinson, E.C., Schuh, A., Wright, R., Fitzgibbon, S.P.,
116 | Bozek, J., Counsell, S.J., Steinweg, J., Vecchiato, K., Passerat-Palmbach, J.,
117 | Lenz, G., Mortari, F., Tenev, T., Duff, E.P., Bastiani, M., Cordero-Grande,
118 | L., Hughes, E., Tusor, N., Tournier, J.-D., Hutter, J., Price, A.N., Teixeira,
119 | R.P.A.G., Murgasova, M., Victor, S., Kelly, C., Rutherford, M.A., Smith, S.,
120 | Edwards, A.D., Hajnal, J.V., Jenkinson, M., Rueckert, D. **The Developing
121 | Human Connectome Project: a Minimal Processing Pipeline for Neonatal
122 | Cortical Surface Reconstruction** *NeuroImage (2018), 173: 88-112.* [DOI:
123 | 10.1016/j.neuroimage.2018.01.054](https://doi.org/10.1016/j.neuroimage.2018.01.054)
124 |
125 | **Surface templates:** Cortical surfaces atlases have been generated
126 | following the procedures described in this paper:
127 |
128 | Bozek, J., Makropoulos, A., Schuh, A., Fitzgibbon, S., Wright, R., Glasser,
129 | M. F., Coalson, T. S., O'Muircheartaigh, J., Hutter, J., Price, A. N.,
130 | Cordero-Grande, L., Teixeira, R. P. A. G., Hughes, E., Tusor, N., Pegoretti
131 | Baruteau, K., Rutherford, M. A., Edwards, A. D., Hajnal, J. V. Smith,
132 | S. M., Rueckert, D., Jenkinson, M., and Robinson, E. C. **Construction of
133 | a neonatal cortical surface atlas using Multimodal Surface Matching in the
134 | Developing Human Connectome Project** *NeuroImage (2018), 179: 11-29.* [DOI:
135 | 10.1016/j.neuroimage.2018.06.018](https://doi.org/10.1016/j.neuroimage.2018.06.018)
136 |
137 | #### Resting-fMRI data processing
138 |
139 | **Automated processing pipeline:** The pipeline described in the following
140 | paper was applied to all dHCP open access fMRI data.
141 |
142 | Fitzgibbon, SP., Harrison, SJ., Jenkinson, M., Baxter, L., Robinson, EC.,
143 | Bastiani, M., Bozek, J., Karolis, V., Cordero Grande, L., Price, AN., Hughes,
144 | E., Makropoulos, A., Passerat-Palmbach, J., Schuh, A., Gao, J., Farahibozorg,
145 | S., O'Muircheartaigh, J., Ciarrusta, J., O'Keeffe, C., Brandon, J., Arichi,
146 | T., Rueckert, D., Hajnal, JV., Edwards, AD., Smith, SM., Duff, E., Andersson,
147 | J. **The developing Human Connectome Project automated functional processing
148 | framework for neonates.**, *NeuroImage (2020), 223: 117303, 2020.* [DOI:
149 | 10.1016/j.neuroimage.2020.117303](https://doi.org/10.1016/j.neuroimage.2020.117303)
150 | *Authors contributed equally.*
151 |
152 | **fMRI motion and distortion correction:** Techniques described in the
153 | following papers were applied to all open access pre-processed fMRI data:
154 |
155 | Andersson, J. L. R., Graham, M. S., Drobnjak, I., Zhang, H.,
156 | and Campbell, J. **Susceptibility-induced distortion that varies
157 | due to motion: Correction in diffusion MR without acquiring
158 | additional data** *NeuroImage (2018), 171: 277–295* [DOI:
159 | 10.1016/j.neuroimage.2017.12.040](https://doi.org/10.1016/j.neuroimage.2017.12.040)
160 |
161 | Andersson, J. L. R., Graham, M. S., Drobnjak, I., Zhang, H.,
162 | Filippini, N., and Bastiani, M. **Towards a comprehensive framework
163 | for movement and distortion correction of diffusion MR images:
164 | Within volume movement** *NeuroImage (2017), 152: 450–466.* [DOI:
165 | 10.1016/j.neuroimage.2017.02.085](https://doi.org/10.1016/j.neuroimage.2017.02.085)
166 |
167 | Andersson, J. L. R., Hutton, C., Ashburner, J., Turner, R.,
168 | and Friston, K. **Modeling Geometric Deformations in EPI
169 | Time Series** *NeuroImage (2001), 13(5): 903–919.* [DOI:
170 | 10.1006/nimg.2001.0746](https://doi.org/10.1006/nimg.2001.0746)
171 |
172 | Andersson, J. L. R., Skare, S., and Ashburner, J. **How to correct
173 | susceptibility distortions in spin-echo echo-planar images: application
174 | to diffusion tensor imaging** *NeuroImage (2003), 20(2): 870–888.* [DOI:
175 | 10.1016/S1053-8119(03)00336-7](https://doi.org/10.1016/S1053-8119(03)00336-7)
176 |
177 | #### Diffusion MRI EDDY pipeline
178 |
179 | **Automated processing pipeline:** The pipeline described in the following
180 | reference was applied to all dHCP open access diffusion data:
181 |
182 | Bastiani, M., Andersson, J.L.R., Cordero-Grande, L., Murgasova, M., Hutter,
183 | J., Price, A.N., Makropoulos, A., Fitzgibbon, S.P., Hughes, E., Rueckert,
184 | D., Suresh, V., Rutherford, M., Edwards, A.D., Smith, S., Tournier,
185 | J. D., Hajnal, J.V., Jbabdi, S., & Sotiropoulos, S.N. **Automated
186 | processing pipeline for neonatal diffusion MRI in the developing
187 | Human Connectome Project** *NeuroImage (2018), 185: 750-763.* [DOI:
188 | 10.1016/j.neuroimage.2018.05.064](https://doi.org/10.1016/j.neuroimage.2018.05.064)
189 |
190 | The pipeline can be downloaded from:
191 |
192 | [https://git.fmrib.ox.ac.uk/matteob/dHCP_neo_dMRI_pipeline_release](https://git.fmrib.ox.ac.uk/matteob/dHCP_neo_dMRI_pipeline_release)
193 |
194 | **Diffusion imaging distortion correction and quality control:** Techniques
195 | described in the following references were applied to all dHCP open access
196 | pre-processed diffusion data; the last reference describes the automated
197 | quality control framework that was used to detect processing issues or
198 | inconsistencies:
199 |
200 | Andersson, J.L.R., Skare, S., and Ashburner, J. **How to correct
201 | susceptibility distortions in spin-echo echo-planar images: application
202 | to diffusion tensor imaging** *NeuroImage (2003), 20: 870-888.* [DOI:
203 | 10.1016/S1053-8119(03)00336-7](https://doi.org/10.1016/S1053-8119(03)00336-7)
204 |
205 | Andersson, J.L.R., and Sotiropoulos, S.N. **An integrated approach
206 | to correction for off-resonance effects and subject movement in
207 | diffusion MR imaging** *NeuroImage (2016), 125: 1063-1078.* [DOI:
208 | 10.1016/j.neuroimage.2015.10.019](https://doi.org/10.1016/j.neuroimage.2015.10.019)
209 |
210 | Andersson, J.L.R., Graham, M.S., Zsoldos, E., and Sotiropoulos,
211 | S.N. **Incorporating outlier detection and replacement into a
212 | non-parametric framework for movement and distortion correction
213 | of diffusion MR images** *NeuroImage (2016), 141: 556-572.* [DOI:
214 | 10.1016/j.neuroimage.2016.06.058](https://doi.org/10.1016/j.neuroimage.2016.06.058)
215 |
216 | Andersson, J.L.R., Graham, M.S., Drobnjak, I., Zhang, H., Filippini,
217 | N., and Bastiani, M. **Towards a comprehensive framework for
218 | movement and distortion correction of diffusion MR images:
219 | Within volume movement** *NeuroImage (2017), 152: 450-466.* [DOI:
220 | 10.1016/j.neuroimage.2017.02.085](https://doi.org/10.1016/j.neuroimage.2017.02.085)
221 |
222 | Andersson, J.L.R., Graham, M.S., Drobnjak, I., Zhang, H., and
223 | Campbell, J. (2018). **Susceptibility-induced distortion that
224 | varies due to motion: Correction in diffusion MR without acquiring
225 | additional data** *NeuroImage (2018), 171: 277-295.* [DOI:
226 | 10.1016/j.neuroimage.2017.12.040](https://doi.org/10.1016/j.neuroimage.2017.12.040)
227 |
228 | Bastiani, M., Cottaar, M., Fitzgibbon, S.P., Suri, S.,
229 | Alfaro-Almagro, F., Sotiropoulos, S.N., Jbabdi, S., & Andersson,
230 | J.L.R. **Automated quality control for within and between studies
231 | diffusion MRI data using a non-parametric framework for movement
232 | and distortion correction** *NeuroImage (2018), 184: 801-812.* [DOI:
233 | 10.1016/j.neuroimage.2018.09.073](https://doi.org/10.1016/j.neuroimage.2018.09.073)
234 |
235 | #### Diffusion MRI SHARD pipeline
236 |
237 | The second dMRI processing pipeline is described in:
238 |
239 | Christiaens, D., Cordero-Grande, L., Pietsch, M., Hutter, J., Price, A.N.,
240 | Hughes, E.J., Vecchiato, K., Deprez, M., Edwards, A.D., Hajnal, J.V., &
241 | Tournier, J-D. **Scattered slice SHARD reconstruction for motion correction
242 | in multi-shell diffusion MRI** *NeuroImage (2021), 225: 117437.* [DOI:
243 | 10.1016/j.neuroimage.2020.117437](https://doi.org/10.1016/j.neuroimage.2020.117437)
244 |
245 | Additionally, inter-slice intensity inconsistencies were corrected with
246 |
247 | Pietsch, M. and Christiaens, D. and Hajnal, J.V. & Tournier,
248 | J-D. **dStripe: slice artefact correction in diffusion MRI
249 | via constrained neural network** *biorxiv (2020)* [DOI:
250 | 10.1101/2020.10.20.347518](https://doi.org/10.1101/2020.10.20.347518)
251 |
--------------------------------------------------------------------------------
/structure.md:
--------------------------------------------------------------------------------
1 | ---
2 | ---
3 |
4 | ## Data directory structure
5 |
6 | The structure of data directory and convention of files naming will follow
7 | the BIDS specification (v1.5.0-dev).
8 |
9 |
10 |
11 | ### Participants and sessions
12 |
13 | The `participants.tsv` for each BIDS pipeline has a number of extra columns
14 | beyond `participant_id`. These have the following meaning:
15 |
16 | Field | Meaning
17 | :------------- | :------
18 | `sex` | Male / Female
19 | `birth_age` | Gestational age at birth in weeks
20 | `birth_weight` | Birthweight (kg)
21 | `singleton` | Singleton pregnancy (S) / Multiple pregnancy (M)
22 |
23 | The `sessions.tsv` file has extra columns beyond `session_id`. These have
24 | the following meaning:
25 |
26 | Field | Meaning
27 | :------------------------ | :------
28 | `scan_age` | Gestational age at scan in weeks
29 | `scan_head_circumference` | Head circumference (cm)
30 | `radiology_score` | Subject status, see below
31 | `scan_number` | 1 for the first scan, 2 for the second
32 | `sedation` | 1 if the subject was sedated during the scan, 0 otherwise
33 |
34 | The MRI scans were reviewed by a specialist perinatal neuroradiologist who
35 | scored each subject using the following scale:
36 |
37 | 1. Normal appearance for age
38 |
39 | 2. Incidental findings with unlikely significance for clinical outcome
40 | or analysis (e.g. subdural haemorrhage. Isolated subependymal cysts. Mild
41 | inferior vermis rotation)
42 |
43 | 3. Incidental findings with unlikely clinical significance but possible
44 | analysis significance (e.g. several punctate lesions or other focal white
45 | matter / cortical lesions not thought to be of clinical significance)
46 |
47 | 4. Incidental findings with possible clinical significance. Unlikely analysis
48 | significance (e.g. Isolated non brain anomaly for example in pituitary /
49 | on tongue)
50 |
51 | 5. Incidental finding with possible / likely significance for both clinical and imaging analysis (e.g. Major lesions within white matter cortex, cerebellum and or basal ganglia; small head / brain < 1st centile)
52 |
53 | 6. `Q`, meaning poor quality anatomical data
54 |
55 | ### Image Spaces
56 |
57 | Images are released in their own native image space with no correspondence to other images (within or across subjects). Transforms (affine and warps) are provided to align images within subjects and to a common space.
58 |
59 | Transform filenames contain a `from` key and a `to` key to indicate the direction of the transform. The `from` value is the space-label for the *origin/source/moving* image, and the `to` value is the space-label for the *reference/target* image.
60 |
61 | Volumetric transforms have the key-value `mode-image`, whereas surface transforms have the key-value `mode-sphere`.
62 |
63 | Non-linear volumetric transforms are FSL format with the extension `.nii.gz`, rigid/affine volumetric transforms have the extension `.mat`, and surface transforms have the extension `.surf.gii`.
64 |
65 | **Examples:**
66 | ```
67 | # volumetric affine transform
68 | xfm/sub-{subid}_ses-{sesid}_from-{space-label}_to-{space-label}_mode-image.mat
69 |
70 | # volumetric nonlinear transform
71 | xfm/sub-{subid}_ses-{sesid}_from-{space-label}_to-{space-label}_mode-image.nii.gz
72 |
73 | # surface transform
74 | xfm/sub-{subid}_ses-{sesid}_hemi-{hemi}_from-{space-label}_to-{space-label}_dens-32k_mode-sphere.surf.gii
75 | ```
76 |
77 | **Space labels:**
78 |
79 | | Label | Description |
80 | | --- | --- |
81 | | `dhcp{age}wk` | [Schuh et al. (2018)](https://gin.g-node.org/BioMedIA/dhcp-volumetric-atlas-groupwise) dHCP atlas week-`{age}` volumetric template space |
82 | | `extdhcp{age}wk` | [Schuh et al. (2018)](https://git.fmrib.ox.ac.uk/seanf/dhcp-resources/-/blob/master/docs/dhcp-augmented-volumetric-atlas-extended.md) dHCP **extended** atlas week-`{age}` volumetric template space |
83 | | `serag{age}wk` | [Serag et al. (2012)](https://brain-development.org/brain-atlases/neonatal-brain-atlases/neonatal-brain-atlas-serag/) atlas week-`{age}` volumetric template space |
84 | | `dhcpSym{age}` | [Bozek et al. (2018)](https://brain-development.org/brain-atlases/atlases-from-the-dhcp-project/cortical-surface-template/) dHCP **symmetric** surface atlas week-`{age}` |
85 | | `T2w` | native structural T2w space (per subject) |
86 | | `bold` | motion and distortion corrected functional space (per subject) |
87 | | `dwi` | motion and distortion corrected diffusion space (per subject) |
88 |
89 | ### Reconstruction Pipeline
90 |
91 | **Path:** `rawdata/sub-{subid}/ses-{sesid}`
92 |
93 | | Group | Description | Filename |
94 | |:--------|:----------------------------------------------------------------------------------------|:--------------------------------------------------------------------------|
95 | | T1 | T1w magnitude image (native acquired stack) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_T1w.nii` |
96 | | T1 | T1w phase image (native acquired stack) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-phase_T1w.nii` |
97 | | T1 | T1w magnitude image (motion corrected and super resolved) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-mcsr_T1w.nii` |
98 | | T1 | T1w phase image (motion corrected and super resolved) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-mcsrphase_T1w.nii` |
99 | | T1 | T1w magnitude image (motion corrected) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-mc_T1w.nii` |
100 | | T1 | T1w phase image (motion corrected) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-mcphase_T1w.nii` |
101 | | T1 | T1w image (combined Slice-to-Volume reconstruction) | `anat/sub-{subid}_ses-{sesid}_rec-SVR_T1w.nii` |
102 | | T2 | T2w magnitude image (native acquired stack) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_T2w.nii` |
103 | | T2 | T2w phase image (native acquired stack) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-phase_T2w.nii` |
104 | | T2 | T2w magnitude image (motion corrected and super resolved) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-mcsr_T2w.nii` |
105 | | T2 | T2w phase image (motion corrected and super resolved) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-mcsrphase_T2w.nii` |
106 | | T2 | T2w magnitude image (motion corrected) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-mc_T2w.nii` |
107 | | T2 | T2w phase image (motion corrected) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-mcphase_T2w.nii` |
108 | | T2 | T2w image (combined Slice-to-Volume reconstruction) | `anat/sub-{subid}_ses-{sesid}_rec-SVR_T2w.nii` |
109 | | T13D | T1w magnitude image (3D MPRAGE) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_acq-MPRAGE_T1w.nii` |
110 | | T13D | T1w phase image (3D MPRAGE) | `anat/sub-{subid}_ses-{sesid}_run-{seqnum}_acq-MPRAGE_rec-phase_T1w.nii` |
111 | | dMRI | Single-band Ref (magnitude) | `dwi/sub-{subid}_ses-{sesid}_run-{seqnum}_sbref.nii` |
112 | | dMRI | Single-band Ref (phase) | `dwi/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-phase_sbref.nii` |
113 | | dMRI | Multi-band dMRI EPI (magnitude) | `dwi/sub-{subid}_ses-{sesid}_run-{seqnum}_dwi.nii` |
114 | | dMRI | Multi-band dMRI EPI (phase) | `dwi/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-phase_dwi.nii` |
115 | | dMRI | Multi-band dMRI EPI (denoised magnitude) | `dwi/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-denoised_dwi.nii` |
116 | | dMRI | Multi-band dMRI EPI (denoised phase) | `dwi/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-denoisedphase_dwi.nii` |
117 | | dMRI | Multi-band dMRI EPI (Chi2-maps) | `dwi/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-chi2_dwi.nii` |
118 | | dMRI | Estimate of error in denoised data | `dwi/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-denoisederror_dwi.nii` |
119 | | fMRI | Single-band Ref (magnitude) | `func/sub-{subid}_ses-{sesid}_run-{seqnum}_task-rest_sbref.nii` |
120 | | fMRI | Single-band Ref (phase) | `func/sub-{subid}_ses-{sesid}_run-{seqnum}_task-rest_rec-phase_sbref.nii` |
121 | | fMRI | Resting fMRI (magnitude) | `func/sub-{subid}_ses-{sesid}_run-{seqnum}_task-rest_bold.nii` |
122 | | fMRI | Resting fMRI (phase) | `func/sub-{subid}_ses-{sesid}_run-{seqnum}_task-rest_phase.nii` |
123 | | fMRI | 4D Spin Echo EPI with different phase encode directions (for topup fieldmap estimation) | `fmap/sub-{subid}_ses-{sesid}_run-{seqnum}_epi.nii` |
124 | | fMRI | 4D Spin Echo EPI (phase) | `fmap/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-phase_epi.nii` |
125 | | B0 | Dual echo-time (magnitude) | `fmap/sub-{subid}_ses-{sesid}_run-{seqnum}_magnitude.nii` |
126 | | B0 | Dual echo-time (phase) | `fmap/sub-{subid}_ses-{sesid}_run-{seqnum}_phase.nii` |
127 | | B0 | B0 field map (Hz) - filtered and smoothed | `fmap/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-filtered_fieldmap.nii` |
128 | | B0 | B0 field map (Hz) - unfiltered | `fmap/sub-{subid}_ses-{sesid}_run-{seqnum}_rec-raw_fieldmap.nii` |
129 | | B1 | DREAM (magnitude) | `B1/sub-{subid}_ses-{sesid}_run-{seqnum}_magnitude.nii` |
130 | | B1 | DREAM (phase) | `B1/sub-{subid}_ses-{sesid}_run-{seqnum}_phase.nii` |
131 | | B1 | B1+ field map (rel. nom. flip) | `B1/sub-{subid}_ses-{sesid}_run-{seqnum}_b1map.nii` |
132 | | recon03 | Multi-band dMRI EPI - release 2 reconstruction | `dwi/sub-{subid}_ses-{sesid}_rec-release2_dwi.nii` |
133 |
134 | **Path:** `sourcedata/sub-{subid}/ses-{sesid}`
135 |
136 | | Group | Description | Filename |
137 | |:--------|:---------------------------------|:-----------------------------------------------------------------|
138 | | fMRI | Resting fMRI Physlog (uncropped) | `func/sub-{subid}_ses-{sesid}_run-{seqnum}_task-rest_physio.log` |
139 |
140 | ### Structural pipeline
141 |
142 | **Path:** `derivatives/dhcp_anat_pipeline/sub-{subid}/ses-{sesid}`
143 |
144 | | Description | Filename |
145 | |:-------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------|
146 | | FSL BET brain mask | `anat/sub-{subid}_ses-{sesid}_desc-brain_mask.nii.gz` |
147 | | Draw-EM regional segmentation (87 labels) | `anat/sub-{subid}_ses-{sesid}_desc-drawem87_dseg.nii.gz` |
148 | | Draw-EM tissue segmentation (9 labels) | `anat/sub-{subid}_ses-{sesid}_desc-drawem9_dseg.nii.gz` |
149 | | Cortical ribbon | `anat/sub-{subid}_ses-{sesid}_desc-ribbon_dseg.nii.gz` |
150 | | T1 weighted image (in T2 space) | `anat/sub-{subid}_ses-{sesid}_T1w.nii.gz` |
151 | | T1 bias field (in T2 space) | `anat/sub-{subid}_ses-{sesid}_desc-biasfield_T1w.nii.gz` |
152 | | T1 weighted, bias corrected image (in T2 space) | `anat/sub-{subid}_ses-{sesid}_desc-restore_T1w.nii.gz` |
153 | | T2 weighted image | `anat/sub-{subid}_ses-{sesid}_T2w.nii.gz` |
154 | | T2 bias field | `anat/sub-{subid}_ses-{sesid}_desc-biasfield_T2w.nii.gz` |
155 | | T2 weighted, bias corrected image | `anat/sub-{subid}_ses-{sesid}_desc-restore_T2w.nii.gz` |
156 | | QC report | `sub-{subid}_ses-{sesid}_qc.pdf` |
157 | | Left/Right white surface | `anat/sub-{subid}_ses-{sesid}_hemi-{hemi}_wm.surf.gii` |
158 | | Left/Right pial surface | `anat/sub-{subid}_ses-{sesid}_hemi-{hemi}_pial.surf.gii` |
159 | | Left/Right mid-thickness surface | `anat/sub-{subid}_ses-{sesid}_hemi-{hemi}_midthickness.surf.gii` |
160 | | Left/Right inflated surface | `anat/sub-{subid}_ses-{sesid}_hemi-{hemi}_inflated.surf.gii` |
161 | | Left/Right very inflated surface | `anat/sub-{subid}_ses-{sesid}_hemi-{hemi}_vinflated.surf.gii` |
162 | | Left/Right spherical surface | `anat/sub-{subid}_ses-{sesid}_hemi-{hemi}_sphere.surf.gii` |
163 | | Cortical curvature | `anat/sub-{subid}_ses-{sesid}_curv.dscalar.nii` |
164 | | Left/Right cortical curvature | `anat/sub-{subid}_ses-{sesid}_hemi-{hemi}_curv.shape.gii` |
165 | | Sulcal depth | `anat/sub-{subid}_ses-{sesid}_sulc.dscalar.nii` |
166 | | Left/Right sulcal depth | `anat/sub-{subid}_ses-{sesid}_hemi-{hemi}_sulc.shape.gii` |
167 | | Cortical thickness | `anat/sub-{subid}_ses-{sesid}_thickness.dscalar.nii` |
168 | | Left/Right cortical thickness | `anat/sub-{subid}_ses-{sesid}_hemi-{hemi}_thickness.shape.gii` |
169 | | Cortical thickness (curvature regressed out) | `anat/sub-{subid}_ses-{sesid}_desc-corr_thickness.dscalar.nii` |
170 | | Left/Right cortical thickness (curvature regressed out) | `anat/sub-{subid}_ses-{sesid}_hemi-{hemi}_desc-corr_thickness.shape.gii` |
171 | | Cortical myelin | `anat/sub-{subid}_ses-{sesid}_myelinmap.dscalar.nii` |
172 | | Left/Right cortical myelin | `anat/sub-{subid}_ses-{sesid}_hemi-{hemi}_myelinmap.shape.gii` |
173 | | Smoothed cortical myelin | `anat/sub-{subid}_ses-{sesid}_desc-smoothed_myelinmap.dscalar.nii` |
174 | | Left/Right smoothed cortical myelin | `anat/sub-{subid}_ses-{sesid}_hemi-{hemi}_desc-smoothed_myelinmap.shape.gii` |
175 | | Cortical regional labels projected from volume | `anat/sub-{subid}_ses-{sesid}_desc-drawem_dseg.dlabel.nii` |
176 | | Left/Right cortical regional labels projected from volume | `anat/sub-{subid}_ses-{sesid}_hemi-{hemi}_desc-drawem_dseg.label.gii` |
177 | | Left/Right Medial wall | `anat/sub-{subid}_ses-{sesid}_hemi-{hemi}_desc-medialwall_mask.shape.gii` |
178 | | Workbench file for loading surfaces | `anat/wb.spec` |
179 | | Warp from native structural space to the subject’s age respective template space | `xfm/sub-{subid}_ses-{sesid}_from-T2w_to-serag{age}wk_mode-image.nii.gz` |
180 | | Warp from the subject’s age respective template space to the native structural space | `xfm/sub-{subid}_ses-{sesid}_from-serag{age}wk_to-T2w_mode-image.nii.gz` |
181 | | Warp from native structural space to the 40-week Serag template space | `xfm/sub-{subid}_ses-{sesid}_from-T2w_to-serag40wk_mode-image.nii.gz` |
182 | | Warp from the 40-week Serag template space to the native structural space | `xfm/sub-{subid}_ses-{sesid}_from-serag40wk_to-T2w_mode-image.nii.gz` |
183 | | Transform from native surface to the dHCP Symmetric 40week surface template | `xfm/sub-{subid}_ses-{sesid}_hemi-{hemi}_from-native_to-dhcpSym40_dens-32k_mode-sphere.surf.gii` |
184 |
185 | ### Diffusion EDDY pipeline
186 |
187 | **Path:** `derivatives/dhcp_dmri_eddy_pipeline/sub-{subid}/ses-{sesid}`
188 |
189 | | Description | Filename |
190 | |:---------------------------------------------------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------|
191 | | Eddy current, susceptibility-by-motion and motion (within and between volumes) corrected super-resolved 4D volume with outlier rejection and replacement | `dwi/sub-{subid}_ses-{sesid}_desc-preproc_dwi.nii.gz` |
192 | | List of b-values | `dwi/sub-{subid}_ses-{sesid}_desc-preproc_dwi.bval` |
193 | | List of post-EDDY rotated gradient directions | `dwi/sub-{subid}_ses-{sesid}_desc-preproc_dwi.bvec` |
194 | | Brain mask | `dwi/sub-{subid}_ses-{sesid}_desc-brain_mask.nii.gz` |
195 | | Estimated field map | `fmap/sub-{subid}_ses-{sesid}_fieldmap.nii.gz` |
196 | | QC report (JSON) | `qc/sub-{subid}_ses-{sesid}_qc.json` |
197 | | QC report (PDF) | `qc/sub-{subid}_ses-{sesid}_qc.pdf` |
198 | | Rigid-body transform from structural to diffusion space | `xfm/sub-{subid}_ses-{sesid}_from-T2w_to-dwi_mode-image.mat` |
199 | | Rigid-body transform from diffusion to structural space | `xfm/sub-{subid}_ses-{sesid}_from-dwi_to-T2w_mode-image.mat` |
200 | | Warp from diffusion space to the extended dHCP 40-week template space | `xfm/sub-{subid}_ses-{sesid}_from-dwi_to-extdhcp40wk_mode-image.nii.gz` |
201 | | Warp from the extended dHCP 40-week template space to diffusion space | `xfm/sub-{subid}_ses-{sesid}_from-extdhcp40wk_to-dwi_mode-image.nii.gz` |
202 | | DTI FA map (based on b=1k shell) | `dwi/sub-{subid}_ses-{sesid}_model-DTI_FA.nii.gz` |
203 | | DTI MD map (based on b=1k shell) | `dwi/sub-{subid}_ses-{sesid}_model-DTI_MD.nii.gz` |
204 | | DTI V1 map (based on b=1k shell) | `dwi/sub-{subid}_ses-{sesid}_model-DTI_EVECS.nii.gz` |
205 | | DTI full tensor (based on b=1k shell) | `dwi/sub-{subid}_ses-{sesid}_model-DTI_diffmodel.nii.gz` |
206 | | DKI mean kurtosis map | `dwi/sub-{subid}_ses-{sesid}_model-DKI_MK.nii.gz`
207 |
208 | ### Diffusion SHARD pipeline
209 |
210 | **Path:** `derivatives/dhcp_dmri_shard_pipeline/sub-{subid}/ses-{sesid}`
211 |
212 | | Description | Filename |
213 | |------------------------------------------------------------------------------|-------------------------------------------------------------------------|
214 | | Topup field map | `fmap/sub-{subid}_ses-{sesid}_fieldmap.nii.gz` |
215 | | Brain mask derived from T2w, warped to preprocessed DWI | `dwi/sub-{subid}_ses-{sesid}_desc-brain_mask.nii.gz` |
216 | | Motion corrected output (5D image of multi-shell SH coefficients, x,y,z,b,m) | `dwi/sub-{subid}_ses-{sesid}_desc-shard_mssh.nii.gz` |
217 | | Motion parameters | `dwi/sub-{subid}_ses-{sesid}_desc-shard_motion.txt` |
218 | | Slice weights | `dwi/sub-{subid}_ses-{sesid}_desc-shard_sliceweights.txt` |
219 | | Local weights for fat-shift suppression | `dwi/sub-{subid}_ses-{sesid}_desc-shard_voxelweights.nii.gz` |
220 | | Preprocessed DWI data (denoising, motion correction, and destriping) | `dwi/sub-{subid}_ses-{sesid}_desc-preproc_dwi.nii.gz` |
221 | | List of b-values in FSL format | `dwi/sub-{subid}_ses-{sesid}_desc-preproc_dwi.bval` |
222 | | List of gradient directions in FSL format | `dwi/sub-{subid}_ses-{sesid}_desc-preproc_dwi.bvec` |
223 | | QC report | `qc/sub-{subid}_ses-{sesid}_qc.html` |
224 | | Rigid-body transform from structural to diffusion space | `xfm/sub-{subid}_ses-{sesid}_from-T2w_to-dwi_mode-image.mat` |
225 | | Rigid-body transform from diffusion to structural space | `xfm/sub-{subid}_ses-{sesid}_from-dwi_to-T2w_mode-image.mat` |
226 | | Warp from diffusion space to the extended dHCP 40-week template space | `xfm/sub-{subid}_ses-{sesid}_from-dwi_to-extdhcp40wk_mode-image.nii.gz` |
227 | | Warp from the extended dHCP 40-week template space to diffusion space | `xfm/sub-{subid}_ses-{sesid}_from-extdhcp40wk_to-dwi_mode-image.nii.gz` |
228 |
229 | ### Functional pipeline
230 |
231 | **Path:** `derivatives/dhcp_fmri_pipeline/sub-{subid}/ses-{sesid}`
232 |
233 | | Description | Filename |
234 | |:-------------------------------------------------------------------------------|:------------------------------------------------------------------------------|
235 | | Multi-band EPI, distortion corrected, motion corrected, 4D image | `func/sub-{subid}_ses-{sesid}_task-rest_desc-mcdc_bold.nii.gz` |
236 | | Multi-band EPI, distortion corrected, motion corrected, FIX denoised, 4D image | `func/sub-{subid}_ses-{sesid}_task-rest_desc-preproc_bold.nii.gz` |
237 | | Motion parameters | `func/sub-{subid}_ses-{sesid}_motion.tsv` |
238 | | Brain mask | `func/sub-{subid}_ses-{sesid}_task-rest_desc-brain_mask.nii.gz` |
239 | | Derived fieldmap, magnitude | `fmap/sub-{subid}_ses-{sesid}_magnitude.nii.gz` |
240 | | Derived fieldmap (rad/s) | `fmap/sub-{subid}_ses-{sesid}_fieldmap.nii.gz` |
241 | | QC report | `sub-{subid}_ses-{sesid}_funcqc.html` |
242 | | Rigid-body transform from functional (mcdc) to single-band Ref space | `xfm/sub-{subid}_ses-{sesid}_from-bold_to-sbref_mode-image.mat` |
243 | | Rigid-body transform from single-band Ref space to structural space | `xfm/sub-{subid}_ses-{sesid}_from-sbref_to-T2w_mode-image.mat` |
244 | | Rigid-body transform from functional (mcdc) to structural space | `xfm/sub-{subid}_ses-{sesid}_from-bold_to-T2w_mode-image.mat` |
245 | | Rigid-body transform from field-map to structural space | `xfm/sub-{subid}_ses-{sesid}_from-fieldmap_to-T2w_mode-image.mat` |
246 | | Warp from functional (mcdc) space to the extended dHCP 40-week template space | `xfm/sub-{subid}_ses-{sesid}_from-bold_to-extdhcp40wk_mode-image.nii.gz` |
247 | | Warp from the structural space to the extended dHCP 40-week template space | `xfm/sub-{subid}_ses-{sesid}_from-T2w_to-extdhcp40wk_mode-image.nii.gz` |
248 | | ICA confound regressors | `func/sub-{subid}_ses-{sesid}_task-rest_desc-fixregressors_timeseries.nii.gz` |
249 | | FOV 4D volumetric mask | `func/sub-{subid}_ses-{sesid}_task-rest_desc-fov4d_mask.nii.gz` |
250 |
--------------------------------------------------------------------------------