Disorder-Free Data are All You Need: Inverse Supervised Learning for Broad-Spectrum Head Disorder Detection

Abstract

Collecting and annotating sufficient data containing disorders is crucial for the development of artificial intelligence (AI)-based medical systems. However, preparing data with complete disorder types and adequate annotations is challenging, which limits the ability of existing AI-based medical systems to diagnose specific disorders. In this paper, we introduce a novel AI-based system that achieves accurate and generalizable broad-spectrum disorder detection without requiring any data containing disorders. Specifically, we obtained a training dataset of 21, 429 disorder-free head computed tomography (CT) scans. We then proposed a learning algorithm called Inverse Supervised Learning (ISL), which learns and understands disorder-free samples instead of disorder-contained ones. This approach enables the identification of all types of disorders. The system achieved AUC values of 0.883, 0.868, and 0.866 on retrospective (127 disorder types, 9, 967 scans), prospective (117 disorder types, 3, 054 scans), and cross-center (46 disorder types, 554 scans) datasets, respectively. These results demonstrate that the system can detect far more disorder types than previous AI-based systems. Additionally, the system provides visually understandable clues, and we developed a diagnosis and visualization software for clinical usage based on these advantages. Furthermore, the ISL-based systems achieved AUC values of 0.893 and 0.895 on pulmonary CT and retinal optical coherence tomography (OCT), respectively, demonstrating that ISL can generalize well to non-head and non-CT images.