Prediction of Post-Bath Body Temperature Using Fuzzy Inference Systems with Hydrotherapy Data

Area of Science:

• Physiology
• Biomedical Engineering
• Computational Intelligence

Background:

• Hydrotherapy leverages water's physical properties for therapeutic benefits, influencing physiological responses.
• Body temperature modulation is key in hydrotherapy, impacting circulation, muscle relaxation, and metabolism.
• Improper temperature control in hydrotherapy presents risks, especially for vulnerable populations.

Purpose of the Study:

• To develop and compare computational models for predicting post-hydrotherapy body temperature.
• To assess the efficacy of fuzzy inference systems (FIS) against machine learning models for temperature prediction.
• To enhance the safety and personalization of hydrotherapy through accurate temperature forecasting.

Main Methods:

• Utilized adaptive neuro-fuzzy inference systems, evolutionary fuzzy inference system (EVOFIS), and enhanced Takagi-Sugeno fuzzy systems.
• Compared FIS models with random forest and support vector machine models.
• Employed hydrotherapy-related datasets for model training and validation.

Main Results:

• The evolutionary fuzzy inference system (EVOFIS) demonstrated superior performance in predicting post-bath body temperature.
• EVOFIS particularly excelled in forecasting deep body temperature, a critical indicator of physiological regulation.
• FIS-based models showed potential for non-invasive temperature prediction.

Conclusions:

• Accurate deep-temperature prediction allows proactive management of hyperthermia risk during hydrotherapy.
• EVOFIS and other FIS models offer a pathway to safer hydrotherapy practices for at-risk individuals.
• These findings support the integration of advanced computational models for personalized and safer hydrotherapy applications.