Heterogeneous Graph Learning for Multi-Modal Medical Data Analysis

Routine clinical visits of a patient produce not only image data, but also non-image data containing information regarding the patient, i.e., medical is multi-modal in nature. Such heterogeneous modalities offer different and complementary perspectives on same resulting more accurate decisions when they are properly combined. However, despite its significance, how to effectively fuse into unified framework has received relatively little attention. In this paper, we propose an effective graph-based called HetMed (Heterogeneous Graph Learning for Multi-modal Medical Data Analysis) fusing data. Specifically, construct multiplex network that incorporates multiple types features patients capture complex relationship between systematic way, which leads decisions. Extensive experiments various real-world datasets demonstrate superiority practicality HetMed. The source code available at https://github.com/Sein-Kim/Multimodal-Medical.