Search papers, labs, and topics across Lattice.
This paper introduces AnomExpert, a novel framework for prenatal ultrasound anomaly diagnosis that utilizes case-level supervision to identify and select relevant anatomical planes without the need for plane-level annotations. By employing learnable plane prototypes and a disease-aware sparse selection mechanism, AnomExpert effectively organizes unordered ultrasound images and enhances diagnostic accuracy. Experimental results on a dataset of 3,654 cases demonstrate that AnomExpert outperforms existing multi-instance learning methods, achieving 86.9% accuracy and 84.2% F1-score with a ViT-small backbone, highlighting its potential for improving prenatal anomaly classification.
AnomExpert achieves 86.9% accuracy in prenatal ultrasound anomaly diagnosis by intelligently selecting relevant anatomical planes without requiring detailed annotations.
Life-limiting congenital anomalies require accurate prenatal diagnosis for appropriate clinical decision-making. Prenatal ultrasound (US) examinations involve multiple anatomical planes, and diagnosis depends on identifying anatomical planes and selecting diagnostically relevant planes for each anomaly. Existing automated methods either rely on plane-level annotations or aggregate heterogeneous images without explicitly modeling these diagnostic capabilities. We propose AnomExpert, a prototype-driven framework for prenatal US anomaly diagnosis using only case-level supervision. AnomExpert introduces learnable plane prototypes to organize unordered images into latent representations corresponding to anatomical planes without requiring plane annotations. A disease-aware sparse selection mechanism further selects diagnostically relevant planes for each anomaly. Experiments on a multi-center dataset of 3,654 cases show that AnomExpert consistently outperforms nine representative multi-instance learning methods. Using a ViT-small backbone, it achieves 86.9% accuracy and 84.2% F1-score while maintaining parameter efficiency. These findings indicate that modeling anatomical plane identification and disease-specific plane selection improves weakly supervised multi-plane prenatal US anomaly classification. The code is available at https://github.com/TIanCat/AnomExpert.
