Meeting Banner
Abstract #1297

Towards Reliable Deep Learning: Feature-Based Out-Of-Distribution Detection for Brain Morphometry

Tommaso Di Noto1,2,3, Lina Bacha1,2,3, Keerthi Prabhu M4, Vincent Dunet3, Attapon Jantarato5, Manuela Vaneckova6, Emmanuelle Le Bars7, Nicolas Menjot de Champfleur7, Punith B. Venkategowda4,8, and Bénédicte Maréchal1,2,3
1Advanced Clinical Imaging Technology, Siemens Healthineers International AG, Lausanne, Switzerland, 2LTS5, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland, 3Department of Diagnostic and Interventional Radiology, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland, 4Magnetic Resonance, Siemens Healthcare Pvt. Ltd., Bangalore, India, 5Chulabhorn Hospital, Bangkok, Thailand, 6MR unit, Department of Radiology, First Faculty of Medicine, Charles University and General University Hospital, Prague, Czech Republic, 7Department of Neuroradiology, Hospital and University of Montpellier, Montpellier, France, 8International Institute of Information Technology, Bangalore, India

Synopsis

Keywords: Other AI/ML, AI/ML Software, Reliable AI; Out-Of-Distribution Detection: Safe AI; Deep Learning; Uncertainty Quantification

Motivation: Despite recent advancements in image segmentation, supervised deep learning algorithms struggle to generalize to Out-Of-Distribution data.

Goal(s): Explore Out-of-Distribution Detection (OODD) in the context of volume-based morphometry for 3D-T1w brain images.

Approach: We estimate a training distribution through a patch-based Convolutional-Neural-Network designed for skull-stripping, which extracts essential features from In-Distribution (ID) data. Then, we classify patients (ID vs. OOD) by calculating the distance in feature space between test patches and this established training distribution.

Results: Our OODD method correctly classifies 98% of Test-ID subjects and 86% Far-OOD. However, it misclassifies most Near-OOD scans suggesting that the skull-stripping-network alone is insufficient for all use-cases.

Impact: We experiment feature-based Out-Of-Distribution (OOD) detection to identify problematic scans for which segmentation results might be unreliable. While Near-OOD remains an area of future improvement, our approach is effective for the majority of use cases and adds negligible computation time.

How to access this content:

For one year after publication, abstracts and videos are only open to registrants of this annual meeting. Registrants should use their existing login information. Non-registrant access can be purchased via the ISMRM E-Library.

After one year, current ISMRM & ISMRT members get free access to both the abstracts and videos. Non-members and non-registrants must purchase access via the ISMRM E-Library.

After two years, the meeting proceedings (abstracts) are opened to the public and require no login information. Videos remain behind password for access by members, registrants and E-Library customers.

Click here for more information on becoming a member.

Keywords