Multi-Scale Evaluation of Uncertainty Quantification Techniques for Deep Learning based MRI Segmentation

Benjamin Lambert, Florence Forbes, Alan Tucholka, Senan Doyle, Michel DojatProceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition 2023 · DOI

Abstract

Efforts are required to design Deep Learning models that are not only powerful, but also capable of expressing the certainty of their predictions. We evaluate 3 state-of-the-art techniques for uncertainty quantification : Monte-Carlo Dropout, Deep Ensemble, and Heteroscedastic models. Evaluation is illustrated on a task of automatic segmentation of White-Matter Hyperintensities in T2-weighted FLAIR MRI sequences of Multiple-Sclerosis patients. Analysis is performed at 3 different scales : the voxel, the lesion, and the whole image. Results indicate the superiority of the Heteroscedastic approach, which ranked first in both the uncertainty and segmentation tasks.
BibTeX
@inproceedings{lambert2023multi,
  author = {Benjamin Lambert and Florence Forbes and Alan Tucholka and Senan Doyle and Michel Dojat},
  title = {{Multi-Scale Evaluation of Uncertainty Quantification Techniques for Deep Learning based MRI Segmentation}},
  booktitle = {Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition},
  year = {2023},
  doi = {10.58530/2022/1791},
}

Loading related content...

Figure