A Hybrid Model to Classify Patients with Chronic Obstructive Respiratory Diseases

Area of Science:

• Digital Health
• Artificial Intelligence in Healthcare
• Patient-Centered Care

Background:

• Increasing aging population and age-related diseases necessitate efficient, affordable healthcare solutions.
• Need for patient empowerment, promoting healthy lifestyles and active disease management.
• Existing solutions often lack personalization for diverse patient needs.

Purpose of the Study:

• To define a hybrid user modeling approach for patients with chronic obstructive respiratory diseases (COPD).
• To identify distinct user groups for personalized health empowerment strategies.
• To develop a foundation for technology-assisted coaching solutions.

Main Methods:

• Employed a hybrid modeling approach combining agglomerative hierarchical clustering and decision tree classification.
• Developed a classification model to identify different user groups based on patient data.
• Generated 7 rules to characterize 5 distinct user clusters.

Main Results:

• Achieved a classification model prediction accuracy of 90.4%.
• Identified 5 distinct user clusters representing different patient profiles.
• Characterized user clusters using 7 specific generated rules.

Conclusions:

• The hybrid user modeling approach effectively categorizes COPD patients.
• This enables the development of personalized coaching solutions adapting to individual patient needs.
• Smartphone-based personalized coaching offers a reliable, affordable disease management alternative.