Linear Dynamic Model to Predict Core Temperature of Persons with Spinal Cord Injury

Owing to disturbances in thermoregulation, persons with spinal cord injuries (SCI) often accumulate heat while exercising under high temperatures. In this case, their core temperatures continue to rise, which necessitates the prediction of their core temperatures to warn them about it. To predict core temperatures, we propose a linear dynamic model and conduct experiments under high temperatures to validate the model. Thereafter, we predict core temperatures and compare the predicted values with the measured data. The fitting errors of our model were lower than the measurement resolution of the core thermometer. The individual parameter for heat exchange is significantly different between the persons with SCI and the able-bodied (AB) persons (p <; 0.1). We predicted the core temperature at 10 min after each sampling. Its prediction errors for the persons with SCI were the same as those in a previous study for AB persons. The mean absolute error of the prediction is significantly different between persons with SCI and AB persons (p <; 0.05). Because the temporal changes in the individual parameters of the persons with SCI were smaller than those in the individual parameters of AB persons, it can be said that by using the proposed model, the core temperatures of persons with SCI can be estimated with fewer data than those of AB persons can be. In future work, we will confirm that our proposed method works properly in various experimental procedures.