A Meta-Analysis of Machine Learning Techniques for Predicting Disease Progression in Electronic Health Records

Yanan Wang

doi:10.63313/hmt.9001

Authors

Yanan Wang The University of British Columbia,Vancouver, Canada Author

DOI:

https://doi.org/10.63313/hmt.9001

Keywords:

Disease Progression Prediction, Machine Learning Techniques, Electronic Health Records (EHR), Meta-Analysis, Predictive Modeling

Abstract

Predicting disease progression is crucial for personalized medicine, enabling tailored treatment strategies. Electronic Health Records (EHRs) provide a valu-able data source for predictive modeling, and integrating machine learning (ML) enhances accuracy and clinical utility. This meta-analysis examines ML tech-niques applied to disease progression prediction using EHR data, synthesizing findings from eight studies published in the last five years. Results reveal diverse ML applications, from traditional regression to deep learning, with performance varying by disease type, data quality, and model complexity. While certain tech-niques show superior predictive accuracy in specific conditions, challenges such as data heterogeneity and model interpretability remain. The findings empha-size the need for disease-specific model selection and improved data integration to enhance clinical applicability. This study provides a roadmap for advancing ML-driven predictive models in personalized healthcare.

References

[1] Z. Ahmed, K. Mohamed, S. Zeeshan, and X. Dong, “Artificial intelligence with mul-ti-functional machine learning platform development for better healthcare and precision medicine,” Database J. Biol. Databases Curation, vol. 2020, p. baaa010, Jan. 2020, doi: 10/gg4rjg.

[2] J. C. Ahn, A. Connell, D. A. Simonetto, C. Hughes, and V. H. Shah, “Application of Artificial In-telligence for the Diagnosis and Treatment of Liver Diseases,” Hepatol. Baltim. Md, vol. 73, no. 6, pp. 2546–2563, Jun. 2021, doi: 10/gskzhh.

[3] Y. Ahuja et al., “Leveraging electronic health records data to predict multiple sclerosis dis-ease activity,” Ann. Clin. Transl. Neurol., vol. 8, no. 4, pp. 800–810, Apr. 2021, doi: 10/gtnqzc.

[4] C. Ye et al., “A Real-Time Early Warning System for Monitoring Inpatient Mortality Risk: Prospective Study Using Electronic Medical Record Data,” J. Med. Internet Res., vol. 21, no. 7, p. e13719, Jul. 2019, doi: 10/ghb396.

[5] J. S. Yoon, Y.-E. Kim, E. J. Lee, H. Kim, and T.-W. Kim, “Systemic factors associated with 10- year glaucoma progression in South Korean population: a single center study based on electronic medical records,” Sci. Rep., vol. 13, no. 1, p. 530, Jan. 2023, doi: 10/gtnqz2.

[6] O. Zaballa, A. Pérez, E. Gómez Inhiesto, T. Acaiturri Ayesta, and J. A. Lozano, “Learning the progression patterns of treatments using a probabilistic generative model,” J. Biomed. In-form., vol. 137, p. 104271, Jan. 2023, doi: 10/grxqff.

[7] D. J. Zamanzadeh etal., “Autopopulus: A Novel Framework for Autoencoder Imputation on Large Clinical Datasets,” Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. IEEE Eng. Med. Biol. Soc. Annu. Int. Conf., vol. 2021, pp. 2303–2309,Nov. 2021, doi: 10/gr8ph5.

[8] L. Chan et al., “Derivation and validation of a machine learning risk score using biomarker and electronic patient data to predict progression of diabetic kidney disease,” Diabetologia, vol. 64, no. 7, pp. 1504– 1515, Jul. 2021, doi: 10/gtnqzb.

[9] A. Dagliati et al., “Using topological data analysis and pseudo time series to infer temporal phenotypes from electronic health records,” Artif. Intell. Med., vol. 108, p. 101930, Aug. 2020, doi: 10/gtnqzt.

[10] E. De Brouwer et al., “Longitudinal machine learning modeling of MS patient trajectories improves predictions of disability progression,” Comput. Methods Programs Biomed., vol. 208, September 2021, 106180