Chronic disease diagnosis model based on convolutional neural network and ensemble learning method.

Area of Science:

• Medical Informatics
• Artificial Intelligence
• Computational Biology

Background:

• Chronic diseases are a leading cause of premature death globally.
• Current diagnostic methods for chronic diseases are often time-consuming and expensive.
• Accurate and timely diagnosis is crucial for effective disease management.

Purpose of the Study:

• To propose a novel hybrid method for enhanced chronic disease diagnosis and prediction.
• To improve classification performance and diagnostic accuracy using ensemble learning with CNN.
• To reduce missed diagnoses and misdiagnoses in chronic disease detection.

Main Methods:

• A hybrid model integrating Convolutional Neural Network (CNN) and ensemble learning (Random Forest and AdaBoost).
• Random Forest (RF) was employed as the base classifier to enhance performance.
• AdaBoost was utilized to replace the CNN's Softmax layer, creating accurate base classifiers.

Main Results:

• The proposed hybrid method demonstrated superior diagnostic accuracy compared to traditional methods (CNN, K-Nearest Neighbor, RF).
• Experimental analysis on a real-world Chronic Electronic Medical Records (C-EMRs) dataset validated the method's effectiveness.
• The approach significantly reduced instances of missed diagnosis and misdiagnosis.

Conclusions:

• The developed method offers a promising tool for improving chronic disease diagnosis.
• It can assist clinicians in making informed decisions and developing targeted interventions.
• The study highlights the potential of AI-driven approaches in reducing diagnostic errors.