An integrated model for medical expense system optimization during diagnosis process based on artificial intelligence...

Area of Science:

• Artificial Intelligence in Healthcare
• Medical Informatics
• Health Economics

Background:

• The healthcare industry is rapidly integrating sophisticated AI algorithms.
• Traditional medical expense systems require optimization for improved decision-making.
• Accurate expense estimation and analysis are crucial for both providers and patients.

Purpose of the Study:

• To propose an integrated AI model for optimizing traditional medical expense systems.
• To enhance decision-making for medical staff and patients regarding healthcare costs.
• To provide a novel perspective on total expense estimation and detailed expense analysis.

Main Methods:

• Development of an integrated model with two intelligent modules: SVM-based and SOM-based.
• Comparative analysis of the SVM-based module against back propagation neural networks and random forests for total expense estimation.
• Implementation of a two-stage clustering process within the SOM-based module for detailed expense analysis.

Main Results:

• The SVM-based module demonstrated superior capability in total expense estimation compared to classic AI techniques.
• The SOM-based module successfully generated decision clusters, clarifying the relationship between detailed expenses and patient information.
• The model was validated using real-world data from coronary heart disease diagnosis in a Shanghai hospital.

Conclusions:

• The proposed AI model innovatively optimizes medical expense systems, providing reliable decision-making information for total and detailed expenses.
• The model serves as a user-friendly tool for medical expense control and therapeutic regimen strategy.
• Successful application in coronary heart disease diagnosis highlights the model's practical utility and accuracy.