Medical Profile Model: Scientific and Practical Applications in Healthcare
This addresses the problem of analyzing complex patient data for healthcare professionals, offering incremental improvements in diagnosis and insurance applications.
The paper tackles representation learning for electronic health records by developing a transformer-based model that creates patient embeddings from disease sequences and demographic data, trained on over one million patients. The model shows clear advantages in diagnosis prediction and improves performance metrics in insurance scoring tasks.
The paper researches the problem of representation learning for electronic health records. We present the patient histories as temporal sequences of diseases for which embeddings are learned in an unsupervised setup with a transformer-based neural network model. Additionally the embedding space includes demographic parameters which allow the creation of generalized patient profiles and successful transfer of medical knowledge to other domains. The training of such a medical profile model has been performed on a dataset of more than one million patients. Detailed model analysis and its comparison with the state-of-the-art method show its clear advantage in the diagnosis prediction task. Further, we show two applications based on the developed profile model. First, a novel Harbinger Disease Discovery method allowing to reveal disease associated hypotheses and potentially are beneficial in the design of epidemiological studies. Second, the patient embeddings extracted from the profile model applied to the insurance scoring task allow significant improvement in the performance metrics.