The virtual doctor prescribing the future: Diagnostics with interactive clinical decision support

Area of Science:

• Biomedical Engineering
• Artificial Intelligence in Healthcare
• Digital Health

Background:

• Healthcare professional shortages, especially in rural areas, limit access to timely medical care.
• The need for innovative solutions to bridge the gap in healthcare accessibility and early disease detection.

Purpose of the Study:

• To introduce and evaluate the Virtual Doctor, a self-service cabin for medical data collection and AI-driven analysis.
• To assess the system's efficacy in early risk detection for type 2 diabetes mellitus and skin cancer.
• To gauge the socio-psychological acceptance of the Virtual Doctor among the general public.

Main Methods:

• Development of a distributed medical data acquisition system with non-invasive biomedical sensors.
• Utilization of two pre-trained deep learning models for predicting type 2 diabetes mellitus and skin cancer risk.
• Socio-psychological acceptance evaluation via a survey of 1217 participants across two use cases.

Main Results:

• The AI models demonstrated predictive capabilities with ROC-AUC scores of 0.729 for diabetes and 0.81 for skin cancer.
• User intention models showed significant results (R² ≥ 0.604) for both routine screenings and antibiotic prescriptions.
• The study indicated broad acceptance of the Virtual Doctor system.

Conclusions:

• The Virtual Doctor system offers a viable solution to enhance healthcare accessibility and facilitate early disease risk detection.
• While generally accepted, clear communication is crucial during serious disease screenings to manage user anxiety.
• The findings support the integration of AI-powered, remote health monitoring solutions in underserved areas.