Dual conditioned diffusion models for out-of-distribution detection: application to fetal ultrasound videos

Conference item

Out-of-distribution (OOD) detection is essential to improve the reliability of machine learning models by detecting samples that do not belong to the training distribution. Detecting OOD samples effectively in certain tasks can pose a challenge because of the substantial heterogeneity within the in-distribution (ID), and the high structural similarity between ID and OOD classes. For instance, when detecting heart views in fetal ultrasound videos there is a high structural similarity between the heart and other anatomies such as the abdomen, and large in-distribution variance as a heart has 5 distinct views and structural variations within each view. To detect OOD samples in this context, the resulting model should generalise to the intra-anatomy variations while rejecting similar OOD samples. In this paper, we introduce dual-conditioned diffusion models (DCDM) where we condition the model on in-distribution class information and latent features of the input image for reconstruction-based OOD detection. This constrains the generative manifold of the model to generate images structurally and semantically similar to those within the in-distribution. The proposed model outperforms reference methods with a 12% improvement in accuracy, 22% higher precision, and an 8% better F1 score.

Authors: Mishra, D; Zhao, H; Saha, P; Papageorghiou, AT; Noble, JA

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Springer
• Host title: Medical Image Computing and Computer Assisted Intervention – MICCAI 2023
• Pages: 216-226
• Series: Lecture Notes in Computer Science
• Series number: 14220
• Publication date: 2023-10-01
• Acceptance date: 2023-06-24
• Event title: 26th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2023)
• Event location: Vancouver, Canada
• Event website: https://conferences.miccai.org/2023/en/
• Event start date: 2023-10-08
• Event end date: 2023-10-12
• DOI: 10.1007/978-3-031-43907-0_21
• EISSN: 1611-3349
• ISSN: 0302-9743
• EISBN: 9783031439070
• ISBN: 9783031439063
• Language: English
• Keywords: heart disease; machine learning and artificial intelligence; cardiovascular; biomedical imaging