Zi at CIKM 2025

On 14 November 2025, Zi Cai presented his work “ProtoEHR: An Interpretable Framework for Hierarchical Healthcare Predictions” at the ACM International Conference on Information and Knowledge Management (CIKM 2025), held in Seoul, South Korea. CIKM is one of the leading international forums for research in information retrieval, data management, and knowledge-based systems, bringing together experts from academia and industry to discuss advances in artificial intelligence and data science.

Zi’s paper introduces ProtoEHR, a new framework designed to improve both the accuracy and interpretability of predictive models built from electronic health records (EHRs). Recognising that EHR data are inherently hierarchical, ProtoEHR leverages this structure through hierarchical prototypes learned at multiple levels. The framework also employs large language models (LLMs) to construct a medical knowledge graph, capturing intrinsic similarities among medical codes in an efficient and scalable manner.

This work marks an important milestone for Zi, representing his first publication completed as a visiting researcher at the Institute of Biomedical Engineering, University of Oxford. For more information, please refer to paper.