Fetal brain QC is a tool to facilitate quality annotations of T2w fetal brain MRI images, by creating interactive html-based visual reports from fetal brain scans. It uses a pair of low-resolution (LR) T2w images with corresponding brain masks to provide snapshots of the brain in the three orientations of the acquisition in the subject-space.
The code is based on MRIQC [1], available at https://github.com/nipreps/mriqc.
Note. The current version of Fetal brain QC only works on low-resolution T2w images.
Disclaimer. Fetal brain QC is not intended for clinical use.
fetal_brain_qc was developed in Ubuntu 22.04 and tested for python 3.9.15
To install this repository, first clone it via
git clone [email protected]:Medical-Image-Analysis-Laboratory/fetal_brain_qc.git
and enter into the directory. Create a conda environment using conda env create -f environment.yml
Clone and install fetal_brain_utils using python -m pip install -e .
Download and install MONAIfbs: clone the repository, go into the repository and install it using python -m pip install -e .
Download the pretrained model here, and to add it to fetal_brain_qc/models/MONAIfbs_dynunet_ckpt.pt.
Download the checkpoint pytorch.ckpt from fetal-IQA at [https://zenodo.org/record/7368570]. Rename it to fetal_IQA_pytorch.ckpt and put it into fetal_brain_qc/models.
Download a checkpoint from pl-fetal-brain-assessment from this link. Rename it to FNNDSC_qcnet_ckpt.hdf5 and put it into fetal_brain_qc/models.
Finally, move back to the fetal_brain_qc repository and install fetal_brain_qc using python -m pip install -e .
Fetal brain QC starts from a BIDS dataset (containing NIfTI formatted images), as well as an additional folder containing brain masks.
The recommended workflow is to use qc_run_pipeline
usage: qc_run_pipeline [-h] [--mask-patterns MASK_PATTERNS [MASK_PATTERNS ...]] [--bids-csv BIDS_CSV] [--anonymize-name | --no-anonymize-name] [--randomize | --no-randomize] [--seed SEED]
[--n-reports N_REPORTS] [--n-raters N_RATERS]
bids_dir out_path
Given a `bids_dir`, lists the LR series in the directory and tries to find corresponding masks given by `mask_patterns`. Then, saves all the found pairs of (LR series, masks) in a CSV file at `bids_csv`
positional arguments:
bids_dir BIDS directory containing the LR series.
out_path Path where the reports will be stored.
optional arguments:
-h, --help show this help message and exit
--mask-patterns MASK_PATTERNS [MASK_PATTERNS ...]
List of patterns to find the LR masks corresponding to the LR series. Patterns will be of the form
"sub-{subject}[/ses-{session}][/{datatype}]/sub-{subject}[_ses-{session}][_acq-{acquisition}][_run-{run}]_{suffix}.nii.gz", and the different fields will be substituted based on
the structure of bids_dir. The code will attempt to find the mask corresponding to the first pattern, and if it fails, will attempt the next one, etc. If no masks are found, a
warning message will be displayed for the given subject, session and run. (default: ['/media/tsanchez/tsanchez_data/data/data_anon/derivatives/refined_masks/sub-{subject}[/ses-{
session}][/{datatype}]/sub-{subject}[_ses-{session}][_acq-{acquisition}][_run-{run}]_{suffix}.nii.gz', '/media/tsanchez/tsanchez_data/data/data_anon/derivatives/automated_masks/
sub-{subject}[/ses-{session}][/{datatype}]/sub-{subject}[_ses-{session}][_acq-{acquisition}][_run-{run}]_{suffix}.nii.gz'])
--bids-csv BIDS_CSV CSV file where the list of available LR series and masks is stored. (default: bids_csv.csv)
--anonymize-name, --no-anonymize-name
Whether an anonymized name must be stored along the paths in `out-csv`. This will determine whether the reports will be anonymous in the end. (default: True)
--randomize, --no-randomize
Whether the order of the reports should be randomized. The number of splits will be given by `n-raters` and the number of reports in each split by `n-reports`. The results will
be stored in different subfolders of `out-path` (default: True)
--seed SEED Seed to control the randomization (to be used with randomize=True). (default: 42)
--n-reports N_REPORTS
Number of reports that should be used in the randomized study (to be used with randomize=True). (default: 100)
--n-raters N_RATERS Number of permutations of the data that must be computed (to be used with randomize=True). (default: 3)
--navigation, --no-navigation
Whether the user should be able to freely navigate between reports. This is
disabled for rating, to force user to process reports sequentially.
(default: False)
Remark. This script runs the whole pipeline of fetal brain QC, i.e. listing of BIDS directory and masks -> (anonymization of data) -> report generation (-> randomization of reports) -> index file generation
Each part of the pipeline can also be called individually, as shown below.
qc_list_bids_csv(+ anonymization)
usage: qc_list_bids_csv [-h] [--mask-patterns MASK_PATTERNS [MASK_PATTERNS ...]] [--out-csv OUT_CSV] [--anonymize-name | --no-anonymize-name] bids-dir
Given a `bids_dir`, lists the LR series in the directory and tries to find corresponding masks given by `mask_patterns`. Then, saves all the found pairs of (LR series, masks) in a CSV file at `out_csv`
positional arguments:
bids_dir BIDS directory containing the LR series.
options:
-h, --help show this help message and exit
--mask-patterns MASK_PATTERNS [MASK_PATTERNS ...]
List of patterns to find the LR masks corresponding to the LR series. Patterns will be of the form
"sub-{subject}[/ses-{session}][/{datatype}]/sub-{subject}[_ses-{session}][_acq-{acquisition}][_run-{run}]_{suffix}.nii.gz", and the different fields will be substituted based on
the structure of bids_dir. The code will attempt to find the mask corresponding to the first pattern, and if it fails, will attempt the next one, etc. If no masks are found, a
warning message will be displayed for the given subject, session and run.
--out-csv OUT_CSV CSV file where the list of available LR series and masks is stored.
--anonymize-name, --no-anonymize-name
Whether an anonymized name must be stored along the paths in `out-csv`. This will determine whether the reports will be anonymous in the end. (default: True)
qc_generate_reports
usage: qc_generate_reports [-h] [--add-js | --no-add-js] out-path bids-csv
positional arguments:
out_path Path where the reports will be stored.
bids_csv Path where the bids config csv file is located.
optional arguments:
-h, --help show this help message and exit
--add-js, --no-add-js
Whether some javascript should be added to the report for interaction with the index file. (default: True)
qc_randomize_reports
usage: qc_randomize_reports [-h] [--seed SEED] [--n-reports N_REPORTS] [--n-raters N_RATERS] reports-path out-path
Randomization of the reports located in `reports_path`. By default, the `n-reports` random reports will be sampled and `n-reports` different permutations of these reports will be saved as subfolders of
`out-path` labelled as split_1 to split_<n-raters>. Each folder will contain an `ordering.csv` file with the randomized ordering to be used.
positional arguments:
reports_path Path where the reports are located
out_path Path where the randomized reports will be stored.
optional arguments:
-h, --help show this help message and exit
--seed SEED Seed to control the randomization (default: 42)
--n-reports N_REPORTS
Number of reports that should be used in the randomized study. (default: 100)
--n-raters N_RATERS Number of raters (number of permutations of the data that must be computed). (default: 3)
qc_generate_index
usage: qc_generate_index [-h] [--add-script-to-reports | --no-add-script-to-reports] [--use-ordering-file | --no-use-ordering-file] reports-path [reports-path ...]
positional arguments:
reports_path Path where the reports are located
options:
-h, --help show this help message and exit
--add-script-to-reports, --no-add-script-to-reports
Whether some javascript should be added to the report for interaction with the index file. (default: False)
--use-ordering-file, --no-use-ordering-file
Whether ordering.csv should be used to construct the ordering of index.html. The file should be located in the report-path folder. (default: False)
--navigation, --no-navigation
Whether the user should be able to freely navigate between reports. This is
disabled for rating, to force user to process reports sequentially.
(default: False)
Part of this work is based on MRIQC, which is licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0.
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.
[1] Esteban, Oscar, et al. "MRIQC: Advancing the automatic prediction of image quality in MRI from unseen sites." PloS one 12.9 (2017): e0184661. [2] Ranzini, Marta, et al. "MONAIfbs: MONAI-based fetal brain MRI deep learning segmentation." arXiv preprint arXiv:2103.13314 (2021).
[3] Semi-supervised learning for fetal brain MRI quality assessment with ROI consistency
[4] Gagoski, Borjan, et al. "Automated detection and reacquisition of motion‐degraded images in fetal HASTE imaging at 3 T." Magnetic Resonance in Medicine 87.4 (2022): 1914-1922.
[5] Lala, Sayeri, et al. "A deep learning approach for image quality assessment of fetal brain MRI." Proceedings of the 27th Annual Meeting of ISMRM, Montréal, Québec, Canada. 2019.
[6] https://github.com/FNNDSC/pl-fetal-brain-assessment