A Diabetes Risk Prediction and Assessment System Improved Based on the Framingham Model

Diabetes, as an increasingly prevalent chronic disease, poses a significant challenge to global public health. Therefore, developing accurate early prediction models is crucial for timely intervention and management of diabetes. This study is based on a hybrid algorithm model combining the Framingham method with clinical rule enhancement. The model predicts diabetes by processing information such as BMI, age, and insulin levels of the subjects, applying fundamental statistical knowledge, and integrating clinical rules for further enhancement. Prior to model operation, unreasonable physiological indicators are preprocessed using the KNN imputation method. The accuracy and clinical applicability of the model are validated using ROC curves. The dataset used in this study (https://www.kaggle.com/datasets/zain280/diabeties-dataset/data) is sourced from the open Kaggle platform. This approach maintains the accuracy of model predictions while improving clinical interpretability's, Special Characters, Footnotes, or Math in Paper Title or Abstract.