A Real-Time Computer-Aided Diagnosis System for Coronary Heart Disease Prediction Using Clinical Information.

Area of Science:

• Cardiology
• Medical Informatics
• Machine Learning

Background:

• Accurate early diagnosis of coronary heart disease (CHD) is crucial for effective patient management.
• Existing deep learning models for CHD prediction face challenges with large datasets and long training times.
• Current machine learning models often lack the accuracy and robustness required for clinical application.

Purpose of the Study:

• To design a fast and high-precision intelligent model for predicting coronary heart disease (CHD) using clinical information.
• To overcome the limitations of existing deep learning and machine learning models in CHD prediction.
• To provide a clinically applicable tool for early CHD diagnosis.

Main Methods:

• Utilized five public datasets comprising patient clinical information.
• Applied singular value decomposition for feature extraction after data preprocessing.
• Developed the CHD prediction model using a multilayer perceptron approach with 5-fold cross-validation.

Main Results:

• The proposed machine learning model demonstrated superior performance compared to other models developed in the study.
• Achieved an Area Under the Curve (AUC) of 99.10% on the total dataset.
• Reported high performance metrics including 96.63% accuracy, 96.50% precision, 97.4% recall, and 97.0% F1-score.

Conclusions:

• The developed model offers high precision and efficiency for coronary heart disease (CHD) prediction.
• The model's performance across different datasets suggests significant potential for medical and clinical diagnosis.
• This intelligent model can aid in the early and accurate identification of CHD.