Machine learning for predicting chronic diseases: a systematic review.

Area of Science:

• Medical Informatics
• Computational Biology
• Public Health

Background:

• Chronic diseases pose a significant global health burden.
• Accurate prediction of chronic diseases is crucial for timely intervention and management.
• Machine learning offers advanced computational tools for predictive modeling.

Purpose of the Study:

• To systematically review existing literature on the application of machine learning for predicting chronic diseases.
• To identify commonly used machine learning models and their predictive performance.
• To assess the potential of machine learning in improving clinical decision-making and healthcare organization.

Main Methods:

• Systematic review of studies from five databases.
• Inclusion criteria focused on machine learning models for chronic disease prediction with reported area under the receiver operating characteristic curve (AUC) values.
• Quality assessment of included studies using the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) scale.

Main Results:

• 42 studies were included in the review.
• Reported AUC values ranged from 0.74 to 1, indicating varying degrees of predictive accuracy.
• K-nearest neighbors, Naive Bayes, deep neural networks, and random forest were frequently utilized and high-performing models.

Conclusions:

• Machine learning models demonstrate significant potential in predicting the occurrence, progression, and determinants of individual chronic diseases.
• The findings support the integration of machine learning into clinical practice for enhanced decision support.
• This research provides valuable insights for optimizing healthcare services and resource allocation.