AI-enhanced telemedicine: transforming resource allocation and cost-efficiency analysis via advanced queueing model

Area of Science:

• Artificial Intelligence
• Healthcare Management
• Operations Research

Background:

• Conventional telemedicine queueing systems struggle with dynamic demands and priority-based care.
• Advanced queueing mechanisms are essential for optimizing digital healthcare services.

Purpose of the Study:

• To integrate an advanced queueing model with AI-scheduling using deep reinforcement learning.
• To optimize digital healthcare by dynamically adjusting doctor availability and prioritizing patients based on urgency and arrival time.

Main Methods:

• Utilized Q-learning, a model-free reinforcement learning algorithm, for resource allocation.
• Implemented AI-scheduling to minimize patient wait times and dynamically adjust doctor assignments.
• Conducted a case study at Dhanwantri Hospital and Research Centre (DHRC), Jaipur, analyzing over 5,000 patient records across 10 simulation runs.

Main Results:

• AI scheduling reduced emergency patient wait times by 40% (95% CI [35%, 45%]).
• Stagnant scheduling increased peak-hour wait times by 80%.
• AI-scheduling demonstrated a statistically significant reduction in wait times (p-value = 0.003) and showed positive financial effects, including reduced patient costs and improved resource allocation.

Conclusions:

• AI-enhanced queue management offers a scalable and cost-efficient approach to modern telemedicine.
• The proposed AI-scheduling model significantly improves patient care and operational efficiency in digital healthcare.
• Dynamic adjustment of doctor availability and patient prioritization are key to optimizing telemedicine services.