Medical Profile Model: Scientific and Practical Applications in Healthcare

TL;DR

This paper introduces a transformer-based medical profile model that learns patient representations from electronic health records, enabling improved diagnosis prediction, disease hypothesis generation, and insurance scoring.

Contribution

It presents a novel unsupervised transformer model incorporating demographic data for comprehensive patient profiling and knowledge transfer in healthcare applications.

Findings

Outperforms state-of-the-art in diagnosis prediction

Enables disease hypothesis discovery for epidemiology

Improves insurance scoring performance

Abstract

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.