Portable Ultra-Low Field MRI Deep-Learning Algorithms for White Matter Lesion Segmentation Improve Accuracy and Reflect Clinical Disability in Multiple Sclerosis
Paper
Loaded from Europe PMC by your browser; OSCR does not store it: doi.org · Europe PMC
The paper is loaded when this pane is shown.
The authors' code
Shell · 46 lines · 1.3 KB · Apache-2.0
- #!/usr/bin/env bash
- # safe-label.sh — label wrapper with a denylist.
- #
- # Usage: safe-label.sh <number> <add|remove> <label>
- #
- # Blocks the "ready-for-fix" label so agent workflows cannot apply it and
- # thereby trigger the bug-fix PR workflow. All other labels pass through.
- # The denylist is intentionally small; this is the structural boundary that
- # makes fix-PR creation a human-only decision.
- #
- # GITHUB_REPOSITORY is provided automatically by GitHub Actions.
- set -euo pipefail
- if [[ $# -ne 3 ]]; then
- echo "Usage: $0 <number> <add|remove> <label>" >&2
- exit 2
- fi
- NUMBER="$1"
- ACTION="$2"
- LABEL="$3"
- # Deny ADD of the ready-for-fix label. REMOVE is allowed so the fix
- # workflow can clear the label after opening its PR.
- case "$ACTION:$LABEL" in
- add:ready-for-fix)
- echo "safe-label.sh: refusing to add label 'ready-for-fix' — reserved for maintainers" >&2
- exit 1
- ;;
- esac
- REPO="${GITHUB_REPOSITORY:?GITHUB_REPOSITORY must be set}"
- case "$ACTION" in
- add)
- exec gh issue edit "$NUMBER" --repo "$REPO" --add-label "$LABEL"
- ;;
- remove)
- exec gh issue edit "$NUMBER" --repo "$REPO" --remove-label "$LABEL"
- ;;
- *)
- echo "safe-label.sh: unknown action '$ACTION' (expected add|remove)" >&2
- exit 2
- ;;
- esac
safe-label.sh at commit 1489d74, under Apache-2.0 · at the source
Overview
- Neuroimmunology Branch, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD, USA
- Sidney Kimmel Medical College, Thomas Jefferson University, Philadelphia, PA, USA
- Oxford Centre for Integrative Neuroimaging - University of Oxford, Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, United Kingdom
- Office of Biostatistics, Division of Intramural Research, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD, USA
- Oxford Centre for Integrative Neuroimaging, Oxford Centre for Human Brain Activity - University of Oxford, Department of Psychiatry, University of Oxford, Oxford, United Kingdom
- qMRI Core Facility, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD, USA
Abstract
White matter lesion (WML) identification, assessment, and characterization using magnetic resonance imaging (MRI) are fundamental for diagnosis and monitoring of multiple sclerosis (MS). Portable ultra-low field (pULF) MRI at 64 millitesla (mT) has been shown to visualize WML with at least one dimension greater than 4 mm. An automated WML segmentation tool catered to pULF-MRI can provide standardized and accurate quantitative measurements of WML volume. In this study, we sought to investigate and compare the accuracy of machine-learning (ML) and deep-learning (DL) pULF MRI segmentation tools. Same-day paired pULF (64mT) and high-field (HF, 3T) MRI scans from 84 adults with MS or suspected-MS (mean age±SD: 48±13, 62 females) included T2-FLAIR and T1w images. Reference WML segmentations were manually annotated on pULF T2-FLAIR for all scans, with WML confirmed with registered HF T2-FLAIR. HF reference WML segmentations were created. Four automated segmentation methods were applied to pULF scans: Method for Inter-Modal Segmentation Analysis (MIMoSA), an ML algorithm trained on HF WML masks; WMH-SynthSeg, a convolutional neural network model with flexible segmentation capabilities across field strengths and resolution; nnU-Net, a DL algorithm trained on pULF reference WML masks; and Pseudo-Label Assisted nnU-Net (PLAn), a DL algorithm pre-trained on HF reference WML masks and refined with 64mT reference WML masks. Two models were trained with nnU-Net, one using T2-FLAIR images only (nnU-Net-FL) and one using T1w and T2-FLAIR images (nnU-Net-FL/
DSC scores with pULF reference WML masks from PLAn-FL (DSC mean±SD: 0.50±0.24) outperformed MIMoSA (0.24±0.20, p<0.0001), WMH-SynthSeg (0.30±0.18, p<0.0001), nnU-Net-FL (0.41±0.24, p<0.0001), and nnU-Net-FL/
nnU-Net and PLAn performed best in segmenting WML on pULF-MRI at 64 mT, providing accurate quantitative estimates of WML burden. Moreover, WML volumes estimated by these algorithms were associated with clinical measures of disability, underscoring their utility for reflecting clinical and radiological disease severity. Given pULF-MRI’s mobility and lower cost, these findings highlight its relevance in clinical trials, particularly in involving more participants who face logistical constraints and barriers.
Reproduced under the paper's license (CC0), from the paper cited above.
Repository
Its files are read in the Code ↔ Paper reader above.
MIC-DKFZ/nnUNet
1489d746e084459581ff20382c1241d8cc2eb258, 22 October 2026Availability: 2 checks, the latest on 27 October 2026: the link answers
- 27 October 2026: the link answers
- 26 September 2026: the link answers
237 files
- .github/
scripts/ , Shell, 46 linessafe-label.sh - .github/
scripts/ , Shell, 29 linessafe-pr-review.sh - documentation/
__init__.py , Python, 1 line - documentation/
competitions/ , Python, 1 lineFLARE24/ Task_1/ __init__.py - documentation/
competitions/ , Python, 209 linesFLARE24/ Task_1/ inference_flare_task1.py - documentation/
competitions/ , Python, 1 lineFLARE24/ Task_2/ __init__.py - documentation/
competitions/ , Python, 467 linesFLARE24/ Task_2/ inference_flare_task2.py - documentation/
competitions/ , Python, 1 lineFLARE24/ __init__.py - documentation/
competitions/ , Python, 1 lineToothfairy2/ __init__.py - documentation/
competitions/ , Python, 359 linesToothfairy2/ inference_script_semseg_ only_customInf2.py - documentation/
competitions/ , Python, 1 line__init__.py - nnunetv2/
__init__.py , Python, 1 line - nnunetv2/
batch_running/ , Python, 1 line__init__.py - nnunetv2/
batch_running/ , Python, 1 linebenchmarking/ __init__.py - nnunetv2/
batch_running/ , Python, 41 linesbenchmarking/ generate_benchmarking_co mmands.py - nnunetv2/
batch_running/ , Python, 69 linesbenchmarking/ summarize_benchmark_resu lts.py - nnunetv2/
batch_running/ , Python, 121 linescollect_results_custom_D ecathlon.py - nnunetv2/
batch_running/ , Python, 17 linescollect_results_custom_D ecathlon_2d.py - nnunetv2/
batch_running/ , Python, 111 linesgenerate_lsf_runs_custom Decathlon.py - nnunetv2/
batch_running/ , Shell, 52 linesjobs.sh - nnunetv2/
batch_running/ , Python, 1 linerelease_trainings/ __init__.py - nnunetv2/
batch_running/ , Python, 1 linerelease_trainings/ nnunetv2_v1/ __init__.py - nnunetv2/
batch_running/ , Python, 112 linesrelease_trainings/ nnunetv2_v1/ collect_results.py - nnunetv2/
batch_running/ , Python, 91 linesrelease_trainings/ nnunetv2_v1/ generate_lsf_commands.py - nnunetv2/
configuration.py , Python, 10 lines - nnunetv2/
dataset_conversion/ , Python, 221 linesDataset015_018_RibFrac_R ibSeg.py - nnunetv2/
dataset_conversion/ , Python, 70 linesDataset021_CTAAorta.py - nnunetv2/
dataset_conversion/ , Python, 55 linesDataset023_AbdomenAtlas1 _1Mini.py - nnunetv2/
dataset_conversion/ , Python, 114 linesDataset027_ACDC.py - nnunetv2/
dataset_conversion/ , Python, 110 linesDataset042_BraTS18.py - nnunetv2/
dataset_conversion/ , Python, 110 linesDataset043_BraTS19.py - nnunetv2/
dataset_conversion/ , Python, 212 linesDataset047_Chaos_Challen ge.py - nnunetv2/
dataset_conversion/ , Python, 84 linesDataset073_Fluo_C3DH_A54 9_SIM.py - nnunetv2/
dataset_conversion/ , Python, 198 linesDataset114_MNMs.py - nnunetv2/
dataset_conversion/ , Python, 61 linesDataset115_EMIDEC.py - nnunetv2/
dataset_conversion/ , Python, 196 linesDataset119_ToothFairy2_A ll.py - nnunetv2/
dataset_conversion/ , Python, 86 linesDataset120_RoadSegmentat ion.py - nnunetv2/
dataset_conversion/ , Python, 97 linesDataset137_BraTS21.py - nnunetv2/
dataset_conversion/ , Python, 68 linesDataset218_Amos2022_task 1.py - nnunetv2/
dataset_conversion/ , Python, 64 linesDataset219_Amos2022_task 2.py - nnunetv2/
dataset_conversion/ , Python, 49 linesDataset220_KiTS2023.py - nnunetv2/
dataset_conversion/ , Python, 70 linesDataset221_AutoPETII_202 3.py - nnunetv2/
dataset_conversion/ , Python, 59 linesDataset223_AMOS2022postC hallenge.py - nnunetv2/
dataset_conversion/ , Python, 60 linesDataset224_AbdomenAtlas1 .0.py - nnunetv2/
dataset_conversion/ , Python, 55 linesDataset226_BraTS2024-Bra TS-GLI.py - nnunetv2/
dataset_conversion/ , Python, 54 linesDataset227_TotalSegmenta torMRI.py - nnunetv2/
dataset_conversion/ , Python, 765 linesDataset237_LIDC.py - nnunetv2/
dataset_conversion/ , Python, 176 linesDataset239_QUADRA_HC.py - nnunetv2/
dataset_conversion/ , Python, 431 linesDataset340_InsectBrain.p y - nnunetv2/
dataset_conversion/ , Python, 32 linesDataset987_dummyDataset4 .py - nnunetv2/
dataset_conversion/ , Python, 32 linesDataset989_dummyDataset4 _2.py - nnunetv2/
dataset_conversion/ , Python, 1 line__init__.py - nnunetv2/
dataset_conversion/ , Python, 131 linesconvert_MSD_dataset.py - nnunetv2/
dataset_conversion/ , Python, 53 linesconvert_raw_dataset_from _old_nnunet_format.py - nnunetv2/
dataset_conversion/ , Python, 73 linesdatasets_for_integration _tests/ Dataset996_IntegrationTe st_Hippocampus_regions_i gnore.py - nnunetv2/
dataset_conversion/ , Python, 37 linesdatasets_for_integration _tests/ Dataset997_IntegrationTe st_Hippocampus_regions.p y - nnunetv2/
dataset_conversion/ , Python, 33 linesdatasets_for_integration _tests/ Dataset998_IntegrationTe st_Hippocampus_ignore.py - nnunetv2/
dataset_conversion/ , Python, 27 linesdatasets_for_integration _tests/ Dataset999_IntegrationTe st_Hippocampus.py - nnunetv2/
dataset_conversion/ , Python, 1 linedatasets_for_integration _tests/ __init__.py - nnunetv2/
dataset_conversion/ , Python, 111 linesgenerate_dataset_json.py - nnunetv2/
ensembling/ , Python, 1 line__init__.py - nnunetv2/
ensembling/ , Python, 207 linesensemble.py - nnunetv2/
evaluation/ , Python, 1 line__init__.py - nnunetv2/
evaluation/ , Python, 58 linesaccumulate_cv_results.py - nnunetv2/
evaluation/ , Python, 349 linesevaluate_predictions.py - nnunetv2/
evaluation/ , Python, 339 linesfind_best_configuration. py - nnunetv2/
experiment_planning/ , Python, 1 line__init__.py - nnunetv2/
experiment_planning/ , Python, 1 linedataset_fingerprint/ __init__.py - nnunetv2/
experiment_planning/ , Python, 222 linesdataset_fingerprint/ fingerprint_extractor.py - nnunetv2/
experiment_planning/ , Python, 1 lineexperiment_planners/ __init__.py - nnunetv2/
experiment_planning/ , Python, 603 linesexperiment_planners/ default_experiment_plann er.py - nnunetv2/
experiment_planning/ , Python, 108 linesexperiment_planners/ network_topology.py - nnunetv2/
experiment_planning/ , Python, 1 lineexperiment_planners/ resampling/ __init__.py - nnunetv2/
experiment_planning/ , Python, 54 linesexperiment_planners/ resampling/ planners_no_resampling.p y - nnunetv2/
experiment_planning/ , Python, 199 linesexperiment_planners/ resampling/ resample_with_torch.py - nnunetv2/
experiment_planning/ , Python, 26 linesexperiment_planners/ resencUNet_planner.py - nnunetv2/
experiment_planning/ , Python, 1 lineexperiment_planners/ residual_unets/ __init__.py - nnunetv2/
experiment_planning/ , Python, 291 linesexperiment_planners/ residual_unets/ residual_encoder_unet_pl anners.py - nnunetv2/
experiment_planning/ , Python, 512 lineslike_nnssl.py - nnunetv2/
experiment_planning/ , Python, 171 linesplan_and_preprocess_api. py - nnunetv2/
experiment_planning/ , Python, 235 linesplan_and_preprocess_entr ypoints.py - nnunetv2/
experiment_planning/ , Python, 1 lineplans_for_pretraining/ __init__.py - nnunetv2/
experiment_planning/ , Python, 82 linesplans_for_pretraining/ move_plans_between_datas ets.py - nnunetv2/
experiment_planning/ , Python, 239 linesverify_dataset_integrity .py - nnunetv2/
imageio/ , Python, 1 line__init__.py - nnunetv2/
imageio/ , Python, 107 linesbase_reader_writer.py - nnunetv2/
imageio/ , Python, 81 linesnatural_image_reader_wri ter.py - nnunetv2/
imageio/ , Python, 222 linesnibabel_reader_writer.py - nnunetv2/
imageio/ , Python, 88 linesreader_writer_registry.p y - nnunetv2/
imageio/ , Python, 234 linessimpleitk_reader_writer. py - nnunetv2/
imageio/ , Python, 100 linestif_reader_writer.py - nnunetv2/
inference/ , Python, 197 linesJHU_inference.py - nnunetv2/
inference/ , Python, 1 line__init__.py - nnunetv2/
inference/ , Python, 315 linesdata_iterators.py - nnunetv2/
inference/ , Python, 100 linesexamples.py - nnunetv2/
inference/ , Python, 160 linesexport_prediction.py - nnunetv2/
inference/ , Python, 1,111 linespredict_from_raw_data.py - nnunetv2/
inference/ , Python, 65 linessliding_window_predictio n.py - nnunetv2/
model_sharing/ , Python, 1 line__init__.py - nnunetv2/
model_sharing/ , Python, 64 linesentry_points.py - nnunetv2/
model_sharing/ , Python, 39 linesmodel_download.py - nnunetv2/
model_sharing/ , Python, 126 linesmodel_export.py - nnunetv2/
model_sharing/ , Python, 8 linesmodel_import.py - nnunetv2/
paths.py , Python, 79 lines - nnunetv2/
postprocessing/ , Python, 1 line__init__.py - nnunetv2/
postprocessing/ , Python, 361 linesremove_connected_compone nts.py - nnunetv2/
preprocessing/ , Python, 1 line__init__.py - nnunetv2/
preprocessing/ , Python, 1 linecropping/ __init__.py - nnunetv2/
preprocessing/ , Python, 39 linescropping/ cropping.py - nnunetv2/
preprocessing/ , Python, 1 linenormalization/ __init__.py - nnunetv2/
preprocessing/ , Python, 104 linesnormalization/ default_normalization_sc hemes.py - nnunetv2/
preprocessing/ , Python, 24 linesnormalization/ map_channel_name_to_norm alization.py - nnunetv2/
preprocessing/ , Python, 1 linepreprocessors/ __init__.py - nnunetv2/
preprocessing/ , Python, 492 linespreprocessors/ default_preprocessor.py - nnunetv2/
preprocessing/ , Python, 1 lineresampling/ __init__.py - nnunetv2/
preprocessing/ , Python, 243 linesresampling/ default_resampling.py - nnunetv2/
preprocessing/ , Python, 13 linesresampling/ no_resampling.py - nnunetv2/
preprocessing/ , Python, 419 linesresampling/ resample_torch.py - nnunetv2/
preprocessing/ , Python, 15 linesresampling/ utils.py - nnunetv2/
preprocessing/ , Python, 1 linesampling_locations/ __init__.py - nnunetv2/
preprocessing/ , Python, 171 linessampling_locations/ extract_sampling_locatio ns.py - nnunetv2/
run/ , Python, 1 line__init__.py - nnunetv2/
run/ , Python, 75 linesload_pretrained_weights. py - nnunetv2/
run/ , Python, 362 linesrun_training.py - nnunetv2/
run/ , Python, 252 linesrun_training_from_pretra ined.py - nnunetv2/
tests/ , Python, 1 line__init__.py - nnunetv2/
tests/ , Python, 1 lineintegration_tests/ __init__.py - nnunetv2/
tests/ , Python, 42 linesintegration_tests/ add_lowres_and_cascade.p y - nnunetv2/
tests/ , Python, 18 linesintegration_tests/ cleanup_integration_test .py - nnunetv2/
tests/ , Shell, 10 linesintegration_tests/ lsf_commands.sh - nnunetv2/
tests/ , Shell, 18 linesintegration_tests/ prepare_integration_test s.sh - nnunetv2/
tests/ , Shell, 27 linesintegration_tests/ run_integration_test.sh - nnunetv2/
tests/ , Python, 75 linesintegration_tests/ run_integration_test_bes tconfig_inference.py - nnunetv2/
tests/ , Shell, 1 lineintegration_tests/ run_integration_test_tra iningOnly_DDP.sh - nnunetv2/
tests/ , Python, 46 linesintegration_tests/ run_nnunet_inference.py - nnunetv2/
tests/ , Python, 688 linestest_checkpoints.py - nnunetv2/
tests/ , Python, 31 linestest_compile_helpers.py - nnunetv2/
tests/ , Python, 53 linestest_copy_file_if_newer. py - nnunetv2/
tests/ , Python, 54 linestest_dataset_fingerprint .py - nnunetv2/
tests/ , Python, 75 linestest_do_split.py - nnunetv2/
tests/ , Python, 304 linestest_evaluate_prediction s.py - nnunetv2/
tests/ , Python, 140 linestest_find_objects.py - nnunetv2/
tests/ , Python, 521 linestest_foreground_location s.py - nnunetv2/
tests/ , Python, 148 linestest_natural_image_tiff_ compression.py - nnunetv2/
tests/ , Python, 45 linestest_paths.py - nnunetv2/
tests/ , Python, 844 linestest_resampling.py - nnunetv2/
tests/ , Python, 63 linestest_simpleitk_reorient_ write_seg.py - nnunetv2/
training/ , Python, 1 line__init__.py - nnunetv2/
training/ , Python, 1 linedata_augmentation/ __init__.py - nnunetv2/
training/ , Python, 24 linesdata_augmentation/ compute_initial_patch_si ze.py - nnunetv2/
training/ , Python, 1 linedata_augmentation/ custom_transforms/ __init__.py - nnunetv2/
training/ , Python, 136 linesdata_augmentation/ custom_transforms/ cascade_transforms.py - nnunetv2/
training/ , Python, 55 linesdata_augmentation/ custom_transforms/ deep_supervision_donwsam pling.py - nnunetv2/
training/ , Python, 22 linesdata_augmentation/ custom_transforms/ masking.py - nnunetv2/
training/ , Python, 31 linesdata_augmentation/ custom_transforms/ region_based_training.py - nnunetv2/
training/ , Python, 45 linesdata_augmentation/ custom_transforms/ transforms_for_dummy_2d. py - nnunetv2/
training/ , Python, 1 linedataloading/ __init__.py - nnunetv2/
training/ , Python, 217 linesdataloading/ data_loader.py - nnunetv2/
training/ , Python, 586 linesdataloading/ foreground_locations.py - nnunetv2/
training/ , Python, 530 linesdataloading/ nnunet_dataset.py - nnunetv2/
training/ , Python, 71 linesdataloading/ utils.py - nnunetv2/
training/ , Python, 1 linelogging/ __init__.py - nnunetv2/
training/ , Python, 304 lineslogging/ nnunet_logger.py - nnunetv2/
training/ , Python, 1 lineloss/ __init__.py - nnunetv2/
training/ , Python, 156 linesloss/ compound_losses.py - nnunetv2/
training/ , Python, 29 linesloss/ deep_supervision.py - nnunetv2/
training/ , Python, 200 linesloss/ dice.py - nnunetv2/
training/ , Python, 32 linesloss/ robust_ce_loss.py - nnunetv2/
training/ , Python, 1 linelr_scheduler/ __init__.py - nnunetv2/
training/ , Python, 26 lineslr_scheduler/ polylr.py - nnunetv2/
training/ , Python, 143 lineslr_scheduler/ warmup.py - nnunetv2/
training/ , Python, 1 linennUNetTrainer/ __init__.py - nnunetv2/
training/ , Python, 1,566 linesnnUNetTrainer/ nnUNetTrainer.py - nnunetv2/
training/ , Python, 1 linennUNetTrainer/ pretraining/ __init__.py - nnunetv2/
training/ , Python, 456 linesnnUNetTrainer/ pretraining/ dynamicPretrainedTrainer .py - nnunetv2/
training/ , Python, 766 linesnnUNetTrainer/ pretraining/ pretrainedTrainer.py - nnunetv2/
training/ , Python, 288 linesnnUNetTrainer/ pretraining/ thrp_primusx_finetuning. py - nnunetv2/
training/ , Python, 1 linennUNetTrainer/ primus/ __init__.py - nnunetv2/
training/ , Python, 511 linesnnUNetTrainer/ primus/ primus_trainers.py - nnunetv2/
training/ , Python, 1 linennUNetTrainer/ variants/ __init__.py - nnunetv2/
training/ , Python, 1 linennUNetTrainer/ variants/ benchmarking/ __init__.py - nnunetv2/
training/ , Python, 67 linesnnUNetTrainer/ variants/ benchmarking/ nnUNetTrainerBenchmark_5 epochs.py - nnunetv2/
training/ , Python, 68 linesnnUNetTrainer/ variants/ benchmarking/ nnUNetTrainerBenchmark_5 epochs_noDataLoading.py - nnunetv2/
training/ , Python, 1 linennUNetTrainer/ variants/ competitions/ __init__.py - nnunetv2/
training/ , Python, 5 linesnnUNetTrainer/ variants/ competitions/ aortaseg24.py - nnunetv2/
training/ , Python, 1 linennUNetTrainer/ variants/ data_augmentation/ __init__.py - nnunetv2/
training/ , Python, 390 linesnnUNetTrainer/ variants/ data_augmentation/ nnUNetTrainerDA5.py - nnunetv2/
training/ , Python, 191 linesnnUNetTrainer/ variants/ data_augmentation/ nnUNetTrainerDAOrd0.py - nnunetv2/
training/ , Python, 34 linesnnUNetTrainer/ variants/ data_augmentation/ nnUNetTrainerNoDA.py - nnunetv2/
training/ , Python, 215 linesnnUNetTrainer/ variants/ data_augmentation/ nnUNetTrainerNoMirroring .py - nnunetv2/
training/ , Python, 35 linesnnUNetTrainer/ variants/ data_augmentation/ nnUNetTrainer_noDummy2DD A.py - nnunetv2/
training/ , Python, 1 linennUNetTrainer/ variants/ loss/ __init__.py - nnunetv2/
training/ , Python, 40 linesnnUNetTrainer/ variants/ loss/ nnUNetTrainerCELoss.py - nnunetv2/
training/ , Python, 59 linesnnUNetTrainer/ variants/ loss/ nnUNetTrainerDiceLoss.py - nnunetv2/
training/ , Python, 76 linesnnUNetTrainer/ variants/ loss/ nnUNetTrainerTopkLoss.py - nnunetv2/
training/ , Python, 1 linennUNetTrainer/ variants/ lr_schedule/ __init__.py - nnunetv2/
training/ , Python, 12 linesnnUNetTrainer/ variants/ lr_schedule/ nnUNetTrainerCosAnneal.p y - nnunetv2/
training/ , Python, 133 linesnnUNetTrainer/ variants/ lr_schedule/ nnUNetTrainer_warmup.py - nnunetv2/
training/ , Python, 1 linennUNetTrainer/ variants/ network_architecture/ __init__.py - nnunetv2/
training/ , Python, 29 linesnnUNetTrainer/ variants/ network_architecture/ nnUNetTrainerBN.py - nnunetv2/
training/ , Python, 15 linesnnUNetTrainer/ variants/ network_architecture/ nnUNetTrainerNoDeepSuper vision.py - nnunetv2/
training/ , Python, 1 linennUNetTrainer/ variants/ optimizer/ __init__.py - nnunetv2/
training/ , Python, 58 linesnnUNetTrainer/ variants/ optimizer/ nnUNetTrainerAdam.py - nnunetv2/
training/ , Python, 65 linesnnUNetTrainer/ variants/ optimizer/ nnUNetTrainerAdan.py - nnunetv2/
training/ , Python, 1 linennUNetTrainer/ variants/ sampling/ __init__.py - nnunetv2/
training/ , Python, 57 linesnnUNetTrainer/ variants/ sampling/ nnUNetTrainer_probabilis ticOversampling.py - nnunetv2/
training/ , Python, 1 linennUNetTrainer/ variants/ training_length/ __init__.py - nnunetv2/
training/ , Python, 98 linesnnUNetTrainer/ variants/ training_length/ nnUNetTrainer_Xepochs.py - nnunetv2/
training/ , Python, 59 linesnnUNetTrainer/ variants/ training_length/ nnUNetTrainer_Xepochs_No Mirroring.py - nnunetv2/
utilities/ , Python, 1 line__init__.py - nnunetv2/
utilities/ , Python, 537 linescheckpoints.py - nnunetv2/
utilities/ , Python, 24 linescollate_outputs.py - nnunetv2/
utilities/ , Python, 141 linesconvert_checkpoints.py - nnunetv2/
utilities/ , Python, 16 linescrossval_split.py - nnunetv2/
utilities/ , Python, 75 linesdataset_name_id_conversi on.py - nnunetv2/
utilities/ , Python, 105 linesddp.py - nnunetv2/
utilities/ , Python, 53 linesddp_allgather.py - nnunetv2/
utilities/ , Python, 50 linesdefault_n_proc_DA.py - nnunetv2/
utilities/ , Python, 132 linesfile_path_utilities.py - nnunetv2/
utilities/ , Python, 174 linesfind_class_by_name.py - nnunetv2/
utilities/ , Python, 56 linesfind_objects.py - nnunetv2/
utilities/ , Python, 91 linesget_network_from_plans.p y - nnunetv2/
utilities/ , Python, 89 linesget_network_via_name.py - nnunetv2/
utilities/ , Python, 59 lineshelpers.py - nnunetv2/
utilities/ , Python, 60 linesjson_export.py - nnunetv2/
utilities/ , Python, 1 linelabel_handling/ __init__.py - nnunetv2/
utilities/ , Python, 351 lineslabel_handling/ label_handling.py - nnunetv2/
utilities/ , Python, 252 linesload_weights_utils.py - nnunetv2/
utilities/ , Python, 12 linesnetwork_initialization.p y - nnunetv2/
utilities/ , Python, 279 linesoverlay_plots.py - nnunetv2/
utilities/ , Python, 1 lineplans_handling/ __init__.py - nnunetv2/
utilities/ , Python, 341 linesplans_handling/ plans_handler.py - nnunetv2/
utilities/ , Python, 51 linespool_utils.py - nnunetv2/
utilities/ , Python, 76 linesutils.py - setup.py, Python, 4 lines
- LICENSE, License, 201 lines
- readme.md, Text, 78 lines
Tracing map
Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.
What the map holds:
- 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
- 235 scripts, each with its path and the digest of its content;
- no match between paragraphs and code yet;
- neither the text of the paper nor the code itself.
Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.
Data
No dataset and no data link were found in the paper.
Availability of data and material
Supplemental data, including deidentified clinical data, deidentified imaging data, and deep-learning models (nnU-Net-FL, nnU-Net-FL/
Reproduced under the paper's license (CC0), from the paper cited above.
Versions
The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.
Version 2, 28 September 2026
- Language: n/a → en
- Authors: added Ashley A. Thommana (0009-0004-1117-8987); Corinne A. Donnay (0009-0006-4856-8422); Gina Norato (0000-0003-1324-0217); María I. Gaitán (0000-0003-4455-5641); Daniel S. Reich (0000-0002-2628-4334); removed Ashley A. Thommana; Corinne A. Donnay; Gina Norato; María I. Gaitán; Daniel S. Reich
Version 1, 27 September 2026: the first record
Recorded: type, journal, dates, 8 authors, 8 keywords, 1 funder, 34 references.
Cite
This paper
Thommana, A. A., Donnay, C. A., Norato, G., Gaitán, M. I., Griffanti, L., Nair, G., Reich, D. S., & Okar, S. V. (2026). Portable Ultra-Low Field MRI Deep-Learning Algorithms for White Matter Lesion Segmentation Improve Accuracy and Reflect Clinical Disability in Multiple Sclerosis. medRxiv (preprint). https://
BibTeX
@article{thommana2026por
author = {Thommana, Ashley A. and Donnay, Corinne A. and Norato, Gina and Gaitán, María I. and Griffanti, Ludovica and Nair, Govind and Reich, Daniel S. and Okar, Serhat V.},
title = {{Portable Ultra-Low Field MRI Deep-Learning Algorithms for White Matter Lesion Segmentation Improve Accuracy and Reflect Clinical Disability in Multiple Sclerosis}},
journal = {medRxiv (preprint)},
year = {2026},
month = jul,
publisher = {medRxiv},
doi = {10.64898/
url = {https://
}
RIS
TY - JOUR
AU - Thommana, Ashley A.
AU - Donnay, Corinne A.
AU - Norato, Gina
AU - Gaitán, María I.
AU - Griffanti, Ludovica
AU - Nair, Govind
AU - Reich, Daniel S.
AU - Okar, Serhat V.
TI - Portable Ultra-Low Field MRI Deep-Learning Algorithms for White Matter Lesion Segmentation Improve Accuracy and Reflect Clinical Disability in Multiple Sclerosis
T2 - medRxiv (preprint)
J2 - medRxiv
PY - 2026
DA - 2026/
PB - medRxiv
DO - 10.64898/
UR - https://
LA - en
ER -
CSL-JSON
{
"id": "10.64898/
"type": "article",
"title": "Portable Ultra-Low Field MRI Deep-Learning Algorithms for White Matter Lesion Segmentation Improve Accuracy and Reflect Clinical Disability in Multiple Sclerosis",
"container-title": "medRxiv (preprint)",
"author": [
{
"family": "Thommana",
"given": "Ashley A."
},
{
"family": "Donnay",
"given": "Corinne A."
},
{
"family": "Norato",
"given": "Gina"
},
{
"family": "Gaitán",
"given": "María I."
},
{
"family": "Griffanti",
"given": "Ludovica"
},
{
"family": "Nair",
"given": "Govind"
},
{
"family": "Reich",
"given": "Daniel S."
},
{
"family": "Okar",
"given": "Serhat V."
}
],
"container-title-short":
"DOI": "10.64898/
"publisher": "medRxiv",
"URL": "https://
"language": "en",
"issued": {
"date-parts": [
[
2026,
7,
17
]
]
}
}
The tracing map gets a citation of its own once an author has validated it and it has a DOI.
Similar papers
The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.
- [1] doi:10.1136/jnnp-2025-335884 [code]
- Diffusivity anisotropy signature of slowly expanding lesions predicts progression independent of relapse activity in multiple sclerosis.Journal: Journal of neurology, neurosurgery, and psychiatryIn common: nnU-Net, pydicom, SimpleITK, 11 other tools, multiple sclerosis, structural MRI / diffusion, clinical / translational, 1 reference
- [2] doi:10.3389/fnimg.2023.1252261 [code]
- Pseudo-Label Assisted nnU-Net enables automatic segmentation of 7T MRI from a single acquisition.Journal: Frontiers in neuroimagingIn common: nnU-Net, pydicom, SimpleITK, 11 other tools, multiple sclerosis, structural MRI / diffusion, 1 reference
- [3] doi:10.3390/pathogens13110920
- Radiological Changes in the Spinal Cord and Brain of Patients with HTLV-1-Associated Myelopathy/
Tropical Spastic Paraparesis (HAM/ TSP). Journal: Pathogens (Basel, Switzerland)In common: multiple sclerosis, structural MRI / diffusion, clinical / translational, 1 reference, 3 authors - [4] doi:10.3389/fmed.2026.1875760 [code]
- Adaptive multi-stage domain unlearning for white-matter lesion segmentation.Journal: Frontiers in medicineIn common: nnU-Net, pydicom, SimpleITK, 11 other tools, structural MRI / diffusion, 1 reference
- [5] doi:10.1371/journal.pone.0323398 [code]
- Ratio maps of T1w/
T2w MRI signal intensity do not improve deep-learning segmentation of pediatric brain tumors. Journal: PloS oneIn common: nnU-Net, pydicom, SimpleITK, 11 other tools, structural MRI / diffusion, 1 reference - [6] doi:10.1016/j.media.2023.102789 [code]
- SynthSeg: Segmentation of brain MRI scans of any contrast and resolution without retraining.Journal: Medical image analysisIn common: nnU-Net, pydicom, SimpleITK, 11 other tools, structural MRI / diffusion, 1 reference
- [7] doi:10.3389/fmed.2021.740248 [code]
- A Clinically-Compatible Workflow for Computer-Aided Assessment of Brain Disease Activity in Multiple Sclerosis Patients.Journal: Frontiers in medicineIn common: nnU-Net, pydicom, SimpleITK, 11 other tools, multiple sclerosis, structural MRI / diffusion, clinical / translational
- [8] doi:10.1186/s12880-025-01687-0 [code]
- The impact of cerebral vessels morphological alteration and white matter hyperintensities burden on the one-year risk of ischemic stroke recurrence.Journal: BMC medical imagingIn common: nnU-Net, pydicom, SimpleITK, 11 other tools, structural MRI / diffusion, clinical / translational
- [9] doi:10.1002/hipo.70124 [code]
- Association Between Anterior Hippocampal Gyrification and Episodic Memory Performance in Neurotypical Young Adults.Journal: HippocampusIn common: nnU-Net, pydicom, SimpleITK, 11 other tools, structural MRI / diffusion
- [10] doi:10.1186/s12880-026-02335-x [code]
- Automatic lateral ventricle and choroid plexus segmentation method in infant brain MR images.Journal: BMC medical imagingIn common: nnU-Net, pydicom, SimpleITK, 11 other tools, structural MRI / diffusion
Contribute
The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.
Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.
Claim this paper
Correct its record
Say what each link of this record is, remove the ones that are not the paper's, add the ones that are missing. The correction becomes a new version of the record, in its Versions section.
Validate its tracing map
You validate the map as this page shows it: 1 repository of the authors' code, each at its verified commit and with its license, 235 scripts, and 0 matches between paragraphs and code (see the Code and Map sections). It then receives a DOI on Zenodo, with you (your ORCID iD) and OSCR as its creators; the code itself is not deposited.
The map's fingerprint: sha256:4e9cf7adc9188136…
Add the badge to the code's README
If you are an author or a maintainer of this paper's code, you can add this badge. It says what OSCR knows of the paper (its code, its validated tracing map) and links the code to this page. Paste one of these snippets under the README's title. Keep the two marks around it, so that the badge can be replaced in place and is never added twice. OSCR changes nothing in your repository.
Markdown
<!-- oscr-badge:start -->
[](https://openscicode.org/badge/doi/10.64898/2026.07.15.26357954)
<!-- oscr-badge:end -->reStructuredText
.. oscr-badge:start
.. image:: https://openscicode.org/badge/doi/10.64898/2026.07.15.26357954.svg
:target: https://openscicode.org/badge/doi/10.64898/2026.07.15.26357954
:alt: OSCR
.. oscr-badge:endHTML
<!-- oscr-badge:start -->
<a href="https://openscicode.org/badge/doi/10.64898/2026.07.15.26357954"><img src="https://openscicode.org/badge/doi/10.64898/2026.07.15.26357954.svg" alt="OSCR"></a>
<!-- oscr-badge:end -->On GitHub, open the README, choose Edit (the pencil), paste the snippet under the title, then "Commit changes".
If you choose to, you can have the badge kept up to date for you, with oscr badge --write from the command line or with the GitHub Action in your repository's workflows. Every other form and option (a card, AsciiDoc, LaTeX, shields.io) is on the badge page.
Request its removal
To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).
Discussion, reproductions, activity
Discussion: code errors and mismatches between the code and the paper about this paper and its code, from readers and its authors, beside the paper and its tracing map.
Report a mismatch between the code and the paper · Report a code error (signed in)
Reproductions: reports from readers who ran the authors' code and did not get the paper's result, what they ran, with which environment, commit and data.
Report a reproduction failure (signed in)
Versions of its code: the releases of its code tied to a version of this paper (the preprint, the accepted manuscript, the version of record…), each with the tracing map versioned with it.
Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.
