The article presents Health-LLM, an AI-based model for disease prediction and health management. Traditional health management methods are limited by static data and uniform standards, preventing full integration with individual situations. Health-LLM overcomes these limitations by integrating health reports into a large model, using professional medical expertise to adjust the weighted scores of health characteristics, and employing a semi-automated feature extraction framework. This approach enhances the analytical power of language models and incorporates expert insights to improve the accuracy of disease prediction. The model has been tested on a large number of health reports and has shown to surpass traditional methods in effectiveness.
Publication date: 1 Feb 2024
Project Page: https://github.com/jmyissb/HealthLLM
Paper: https://arxiv.org/pdf/2402.00746
