Deep Bayesian networks for uncertainty estimation and adversarial resistance of white matter hyperintensity segmentation
Open accessThe DOI matches its Crossref record; 15 of the 15 references it cites were found in a registry or an index.
Abstract
White matter hyperintensities (WMHs) are frequently observed on structural neuroimaging of elderly populations and are associated with cognitive decline and increased risk of dementia. Many existing WMH segmentation algorithms produce suboptimal results in populations with vascular lesions or brain atrophy, or require parameter tuning and are computationally expensive. Additionally, most algorithms do not generate a confidence estimate of segmentation quality, limiting their interpretation. MRI‐based segmentation methods are often sensitive to acquisition protocols, scanners, noise‐level, and image contrast, failing to generalize to other populations and out‐of‐distribution datasets. Given these concerns, we propose a novel Bayesian 3D convolutional neural network with a U‐Net architecture that automatically segments WMH, provides uncertainty estimates of the segmentation output for quality control, and is robust to changes in acquisition protocols. We also provide a second model to differentiate deep and periventricular WMH. Four hundred thirty‐two subjects were recruited to train the CNNs from four multisite imaging studies. A separate test set of 158 subjects was used for evaluation, including an unseen multisite study. We compared our model to two established state‐of‐the‐art techniques (BIANCA and DeepMedic), highlighting its accuracy and efficiency. Our Bayesian 3D U‐Net achieved the highest Dice similarity coefficient of 0.89 ± 0.08 and the lowest modified Hausdorff distance of 2.98 ± 4.40 mm. We further validated our models highlighting their robustness on “clinical adversarial cases” simulating data with low signal‐to‐noise ratio, low resolution, and different contrast (stemming from MRI sequences with different parameters). Our pipeline and models are available at: https://hypermapp3r.readthedocs.io .
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
U-Net: Convolutional Networks for Biomedical Image Segmentation
Ronneberger, O., Fischer, P., & Brox, T. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24574-4_28 · Cited by 99,404
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Dropout as a bayesian approximation: Representing model uncertainty in deep learning
Gal, Y., & Ghahramani, Z. (2015) · arXiv · doi:10.48550/arxiv.1506.02142 · Cited by 6,471
Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles
Lakshminarayanan, B., Pritzel, A., & Blundell, C. (2016) · arXiv · doi:10.48550/arxiv.1612.01474 · Cited by 3,666 · cited 1×
“To build a deep learning model that generalizes well, a large and diverse amount of data is required to avoid overfitting (LeCun, Bengio, and Hinton 2015).”
BackgroundEfficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation
Kamnitsas, K., Ledig, C., Newcombe, V. F. J., Simpson, J. P., Kane, A. D., Menon, D. K., Rueckert, D., & Glocker, B. (2017) · Medical Image Analysis · doi:10.1016/j.media.2016.10.004 · Cited by 3,498 · cited 15×
“Our models and DeepMedic had the highest volume correlations with manual WMH compared to the other tested techniques (r = 0.99, p 0.0001), while BIANCA had the lowest volume correlations (r = 0.91, p 0.0001).”
MethodInfluentialWhat Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?
Kendall, A., & Gal, Y. (2017) · arXiv · doi:10.48550/arxiv.1703.04977 · Cited by 3,088
Debette, S., & Markus, H. S. (2010) · BMJ · doi:10.1136/bmj.c3666 · Cited by 2,328 · cited 1×
“WMH have been linked to global cognitive impairment (Jokinen et al. 2020), a decline in mental processing speed (van den Heuvel 2006), increased risk of late onset depression (Herrmann, Le Masurier, and Ebmeier 2008), and increased risk of stroke, dementia, and mortality (Debette and Markus 2010).”
BackgroundKendall, A., Badrinarayanan, V., & Cipolla, R. (2017) · Procedings of the British Machine Vision Conference 2017 · doi:10.5244/c.31.57 · Cited by 784
Zhu, Y., & Zabaras, N. (2018) · Journal of Computational Physics · doi:10.1016/j.jcp.2018.04.018 · Cited by 704 · cited 1×
“Previous works have used Stochastic Variational Gradient Descent (SVGD) to perform approximate Bayesian inference on uncertain CNN parameters (Zhu and Zabaras 2018).”
MethodWang, G., Li, W., Aertsen, M., Deprest, J., Ourselin, S., & Vercauteren, T. (2019) · Neurocomputing · doi:10.1016/j.neucom.2019.01.103 · Cited by 659
Lee, S., Viqar, F., Zimmerman, M. E., Narkhede, A., Tosto, G., Benzinger, T. L. S., Marcus, D. S., Fagan, A. M., Goate, A., Fox, N. C., et al. (2016) · Annals of Neurology · doi:10.1002/ana.24647 · Cited by 542 · cited 1×
“…hyperintensities (WMH) are commonly observed MRI-based biomarkers of cerebral small vessel disease and have been associated with aging and neurodegenerative diseases such as Parkinson’s disease (PD) and Alzheimer’s disease (AD) (S. Lee et al. 2016; Prins and Scheltens 2015; S.-J. Lee et al. 2010).”
BackgroundWhite matter hyperintensities in late life depression: a systematic review
Herrmann, L. L., Le Masurier, M., & Ebmeier, K. P. (2007) · Journal of Neurology, Neurosurgery & Psychiatry · doi:10.1136/jnnp.2007.124651 · Cited by 456
Data Augmentation for Brain-Tumor Segmentation: A Review
Nalepa, J., Marcinkiewicz, M., & Kawulok, M. (2019) · Frontiers in Computational Neuroscience · doi:10.3389/fncom.2019.00083 · Cited by 340
Carmichael, O. (2010) · Archives of Neurology · doi:10.1001/archneurol.2010.284 · Cited by 248
Cited by
Lambert, B., Forbes, F., Doyle, S., Dehaene, H., & Dojat, M. (2024) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2024.102830 · Cited by 269
Deep learning applications in vascular dementia using neuroimaging
Dong, C., & Hayashi, S. (2024) · Current Opinion in Psychiatry · doi:10.1097/yco.0000000000000920 · Cited by 13
Teghipco, A., Newman-Norlund, R., Fridriksson, J., Rorden, C., & Bonilha, L. (2024) · Communications Medicine · doi:10.1038/s43856-024-00541-8 · Cited by 11
Lin, Z., Hong, Y., Hu, Y., Xiao, Q., Peng, S., Wu, D., Guo, T., & Jiang, D. (2025) · Alzheimer's & Dementia · doi:10.1002/alz.70990 · Cited by 3
Metadata
- Authors
- Mojiri Forooshani, P., Biparva, M., Ntiri, E. E., Ramirez, J., Boone, L., Holmes, M. F., Adamo, S., Gao, F., Ozzoude, M., Scott, C. J. M., Dowlatshahi, D., Lawrence‐Dewar, J. M., Kwan, D., Lang, A. E., Marcotte, K., Leonard, C., Rochon, E., Heyn, C., Bartha, R., Strother, S., Tardif, J., Symons, S., Masellis, M., Swartz, R. H., Moody, A., Black, S. E., & Goubran, M.
- Venue
- Human Brain Mapping
- Publisher
- Wiley
- Published
- 2022-01-28
- Volume
- 43
- Issue
- 7
- Pages
- 2089-2108
- ISSN
- 1065-9471 (ISSN-L)
- 1097-0193
- OpenAlex
- W4221129115
- Cited by
- 50
- References
- 15
- FWCI
- 3.72
- Type
- journal article
- Language
- en
- Open access
- gold
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.