DK-BEHRT: Teaching Language Models International Classification of Disease (ICD) Codes using Known Disease Descriptions.

An, U., Lee, S. A., Jeong, M., Gorla, A., Chiang, J. N., and Sankararaman, S.

Proceedings of the First AAAI Bridge Program on AI for Medicine and Healthcare, 2025

A language-model architecture that incorporates biomedical disease-description embeddings into longitudinal EHR models, improving AUPR by up to 62% and AUROC by up to 21% over baseline models.