Development of a Prediction Algorithm for Chronic Disease Using MyData.

Area of Science:

• Computational biology and bioinformatics
• Medical informatics and health data science

Background:

• Chronic diseases like diabetes and hypertension pose significant public health challenges.
• Early detection and personalized prevention are crucial for managing these conditions effectively.
• Analyzing personal health records can reveal patterns predictive of disease onset.

Purpose of the Study:

• To develop and evaluate deep learning models for predicting the onset of diabetes and hypertension.
• To utilize time-series data patterns from personal health records for predictive modeling.
• To leverage the MyData initiative for enhanced personalized prediction and early detection.

Main Methods:

• Application of deep learning algorithms to analyze time-series data from personal health records.
• Utilizing the MyData initiative framework to access and process individual health data.
• Performance evaluation of predictive models using metrics such as Area Under the Receiver Operating Characteristic Curve (AUROC), F1-score, and Recall.

Main Results:

• The developed models demonstrate potential for accurate prediction of chronic disease onset.
• Evaluation metrics (AUROC, F1-score, Recall) indicate the models' predictive capabilities.
• The framework shows promise for personalized prediction of diabetes and hypertension.

Conclusions:

• The study establishes a foundation for using deep learning on personal health records for chronic disease prediction.
• The developed framework can significantly improve preventive care strategies.
• Further generalization of the models is needed for broader real-world clinical applications.