Framework for Personalized Real-Time Control of Hidden Temperature Variables in Therapeutic Knee Cooling

Area of Science:

• Biomedical Engineering
• Rehabilitation Medicine
• Control Systems

Background:

• Cryotherapy is crucial for post-knee surgery recovery.
• Individual patient feedback is vital for effective cryotherapy.
• Current methods lack personalized, real-time temperature control.

Purpose of the Study:

• To develop and evaluate a framework for personalized, real-time control of inner knee temperature during cryotherapy.
• To enable smart, adaptive cooling based on individual patient needs.
• To improve cryotherapy efficacy and safety after knee surgery.

Main Methods:

• A feedback control loop framework was designed.
• Machine learning models predict inner knee temperature using surface temperature data.
• A fuzzy proportional-derivative controller manages cooling temperature.
• Framework evaluated using computer simulations and physiological change scenarios.

Main Results:

• Controlled cooling is essential, particularly for small and large knee sizes due to varying sensitivity.
• The framework effectively adapts to dynamic physiological changes (e.g., blood flow) and altered target temperatures.
• Near real-time control of inner knee temperature was achieved.

Conclusions:

• The developed framework offers a robust and controllable solution for personalized cryotherapy.
• Machine learning-based temperature prediction enhances cryotherapy effectiveness.
• Adaptive control strategies are key to optimizing patient recovery after knee surgery.