A new cloud-based method for composition of healthcare services using deep reinforcement learning and Kalman...

Area of Science:

• Health Informatics
• Cloud Computing
• Artificial Intelligence

Background:

• Cloud computing offers scalable and on-demand services beneficial for healthcare.
• Selecting and composing cloud-based healthcare services is crucial but faces challenges like high energy consumption and response times.

Purpose of the Study:

• To develop a novel layered method for selecting and evaluating cloud-based healthcare services.
• To address issues of energy consumption, cost, and response time in healthcare service composition.

Main Methods:

• A novel layered method integrating Deep Reinforcement Learning (Deep RL), Kalman filtering, and repeated training.
• Evaluation of healthcare service selection and composition solutions.

Main Results:

• The proposed method achieved acceptable results in availability, reliability, energy consumption, and response time.
• Demonstrated improvements compared to existing methods for cloud healthcare service optimization.

Conclusions:

• The developed Deep RL and Kalman filtering-based method offers an effective solution for optimizing cloud healthcare services.
• This approach enhances the efficiency and viability of healthcare services through improved selection and composition.