Medical-Knowledge Driven Multiple Instance Learning for Classifying Severe Abdominal Anomalies on Prenatal Ultrasound

2025-07-02

AI Summary

Prenatal precise diagnosis of severe fetal abdominal malformations faces challenges including heavy reliance on standardized anatomical planes and weak case-level discriminative capability. To address these, we propose the first end-to-end case-level classification framework for fetal abdominal anomalies—eliminating the need for standardized plane localization. Our method (1) formulates ultrasound sequences as bags in a multiple-instance learning paradigm, with individual frames as instances; (2) introduces a Mixture-of-Attention Experts (MoAE) module to dynamically weight contributions from heterogeneous scanning planes; (3) designs Medical-knowledge-guided Frame Selection (MFS), a self-supervised image token selection strategy to enhance feature discriminability; and (4) integrates Prompt-based Prototype Learning (PPL) to improve semantic alignment across anomaly categories. Evaluated on a large-scale prenatal ultrasound dataset comprising 2,419 cases, 24,748 images, and six anomaly classes, our approach significantly outperforms state-of-the-art methods. Code is publicly available.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Computer Vision: Multi-modal Vision

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Fetal abdominal malformations are serious congenital anomalies that require accurate diagnosis to guide pregnancy management and reduce mortality. Although AI has demonstrated significant potential in medical diagnosis, its application to prenatal abdominal anomalies remains limited. Most existing studies focus on image-level classification and rely on standard plane localization, placing less emphasis on case-level diagnosis. In this paper, we develop a case-level multiple instance learning (MIL)-based method, free of standard plane localization, for classifying fetal abdominal anomalies in prenatal ultrasound. Our contribution is three-fold. First, we adopt a mixture-of-attention-experts module (MoAE) to weight different attention heads for various planes. Secondly, we propose a medical-knowledge-driven feature selection module (MFS) to align image features with medical knowledge, performing self-supervised image token selection at the case-level. Finally, we propose a prompt-based prototype learning (PPL) to enhance the MFS. Extensively validated on a large prenatal abdominal ultrasound dataset containing 2,419 cases, with a total of 24,748 images and 6 categories, our proposed method outperforms the state-of-the-art competitors. Codes are available at:https://github.com/LL-AC/AAcls.

Problem

• Classifying severe fetal abdominal anomalies without standard plane localization
• Aligning image features with medical knowledge for case-level diagnosis
• Improving prenatal ultrasound diagnosis accuracy using multiple instance learning

Innovation

• Mixture-of-attention-experts module for plane weighting
• Medical-knowledge-driven feature selection module
• Prompt-based prototype learning to enhance features