Research on Deep Mining of Medical and Health Data and Disease Prediction Based on Machine Learning

DOI: https://doi.org/10.62517/jmhs.202605406

Author(s)

Yuxi Liu

Affiliation(s)

College of St Hild & St Bede, Durham University, Durham, United Kingdom

Abstract

With the rapid advancement of medical informatization, massive multi-source medical and health data have provided an unprecedented data foundation for disease prediction. This paper systematically analyzes the type classification, fusion processing and feature engineering of medical and health data, constructs a disease-prediction model system centered on ensemble learning, and introduces interpretability methods such as SHAP to realize model transparency. Multi-task prediction frameworks are designed for chronic diseases and acute infectious diseases respectively, and model performance is evaluated using metrics including AUC and F1-score together with multi-fold cross-validation. Experimental results show that the ensemble learning method achieves an AUC above 0.90 in both diabetes and pneumonia prediction tasks, approximately 8 % higher than single baseline models. Interpretability outputs effectively assist clinical decision-making. Machine-learning-driven deep mining of medical data possesses remarkable clinical translational value and is expected to advance disease prediction from empirical judgment to precise quantitative decision-making.

Keywords

Machine Learning; Medical and Health Data; Disease Prediction; Ensemble Learning; Interpretability

References

[1] Lu J X, Ren C L, Sun Q. Research on machine-learning optimization applications in big medical-and-health data[J]. Computer Knowledge and Technology, 2026, 22(14): 64-68. [2] Shi X Q, Zhao J, Wang L L, et al. Review on precision-medicine applications based on big-data-analytics technology[J]. Chinese Hospital Management, 2021, 41(5): 26-31. [3] Li T H. Construction of an early-prevention-and-control platform for precise prediction of infectious diseases based on health-medical data from urban medical-alliance systems[J]. China Informatization, 2021, (3): 103-105+102. [4] Cheng W Q, Gao Z S, Liu Y. Research on collaboration and applications of medical-and-health data based on trusted-data spaces[J/OL]. China Digital Medicine:1-8[2026-09-22]. https://link.cnki.net/urlid/11.5550.R.20260911.2214.006. [5] Zhang Y Y, Zou X K, Xian L L. New disease-prevention-and-control modes driven by big health-medical data[J]. China Health Industry, 2024, 21(3): 132-135.