A Cognitive Explainer for Fetal ultrasound images classifier Based on Medical Concepts
This addresses the lack of transparency in AI for medical imaging, offering interpretability to assist inexperienced operators in fetal ultrasound analysis.
The paper tackles the problem of interpreting deep neural networks for fetal ultrasound scan plane detection by proposing a framework that explains decisions using medical concepts, providing clinicians with understandable insights.
Fetal standard scan plane detection during 2-D mid-pregnancy examinations is a highly complex task, which requires extensive medical knowledge and years of training. Although deep neural networks (DNN) can assist inexperienced operators in these tasks, their lack of transparency and interpretability limit their application. Despite some researchers have been committed to visualizing the decision process of DNN, most of them only focus on the pixel-level features and do not take into account the medical prior knowledge. In this work, we propose an interpretable framework based on key medical concepts, which provides explanations from the perspective of clinicians' cognition. Moreover, we utilize a concept-based graph convolutional neural(GCN) network to construct the relationships between key medical concepts. Extensive experimental analysis on a private dataset has shown that the proposed method provides easy-to-understand insights about reasoning results for clinicians.