A new method for disease diagnosis based on hierarchical BRB with power set

Area of Science:

• Medical Informatics
• Artificial Intelligence in Medicine
• Computational Statistics

Background:

• Accurate disease diagnosis is crucial for effective treatment planning.
• Existing belief rule base (BRB) models struggle with combinatorial rule explosion and local ignorance in disease diagnosis.
• Handling uncertainty is a key challenge in developing robust diagnostic models.

Purpose of the Study:

• To propose a novel hierarchical belief rule base with power set (H-BRBp) model for disease diagnosis.
• To overcome the limitations of traditional BRB models in handling uncertainty and local ignorance.
• To enhance the accuracy and efficiency of medical diagnostic systems.

Main Methods:

• Analysis and preprocessing of patient physiological data using Principal Component Regression (PCR).
• Construction of the proposed hierarchical belief rule base with power set (H-BRBp) model.
• Experimental validation using lumbar spine disease diagnosis and comparative analysis.

Main Results:

• The H-BRBp model effectively addresses combinatorial rule explosion and local ignorance in disease diagnosis.
• Demonstrated improved diagnostic accuracy and performance compared to existing methods.
• Successfully applied to lumbar spine disease diagnosis, validating its practical utility.

Conclusions:

• The H-BRBp model offers a significant advancement in uncertainty-based disease diagnosis.
• This approach enhances the reliability and effectiveness of AI-driven medical diagnostic tools.
• The proposed model provides a robust framework for future developments in computational medicine.