A closed-loop hybrid physiological model relating to subjects under physical stress

Area of Science:

• Physiological modeling
• Human systems engineering
• Biomedical engineering

Background:

• Understanding human physiological responses to physical stress is crucial for operator performance in demanding environments.
• Existing models may lack the comprehensive scope to integrate cardiovascular, respiratory, thermoregulatory, and neural activity.
• A detailed physiological model is needed to predict performance breakdown under stress.

Purpose of the Study:

• To develop a comprehensive physiological model of human response to physical stress.
• To describe the behavior of cardiovascular, respiratory, thermoregulation, and brain activity under physical workload.
• To enable prediction of performance breakdown in critical operational settings.

Main Methods:

• Grey-box modeling approach integrating cardiovascular, respiratory, thermoregulation, and EEG signals.
• Experimental setup with a recumbent bicycle for physical load induction.
• Data acquisition using Finapres and ProComp+ monitors for 16 healthy volunteers.

Main Results:

• The model accurately predicted heart rate, blood pressure, body temperature, respiration, and EEG signals via closed-loop simulations.
• Residual analyses confirmed model reliability, with residuals within 90% confidence bands and normally distributed histograms.
• Neural network integration enhanced the model's predictive capabilities across diverse subject dynamics.

Conclusions:

• The developed physiological model effectively predicts the impact of stress on key physiological variables.
• The model can forecast operator performance breakdown in critical environments.
• The architecture supports feedback control for managing stress by defining safe operating ranges for psycho-physiological variables.