Adaptive neuro-fuzzy inference systems with k-fold cross-validation for energy expenditure predictions based on heart rate

This paper presents a new model based on adaptive neuro-fuzzy inference systems (ANFIS) to predict oxygen consumption (VO2) from easily measured variables. The ANFIS prediction model consists of three ANFIS modules for estimating the Flex-HR parameters. Each module was developed based on clustering a training set of data samples relevant to that module and then the ANFIS prediction model was tested against a validation data set. Fifty-eight participants performed the Meyer and Flenghi step-test, during which heart rate (HR) and VO2 were measured. Results indicated no significant difference between observed and estimated Flex-HR parameters and between measured and estimated VO2 in the overall HR range, and separately in different HR ranges. The ANFIS prediction model (MAE=3mlkg^-1min^-1) demonstrated better performance than Rennie etal.'s (MAE=7mlkg^-1min^-1) and Keytel etal.'s (MAE=6mlkg^-1min^-1) models, and comparable performance with the standard Flex-HR method (MAE=2.3mlkg^-1min^-1) throughout the HR range. The ANFIS model thus provides practitioners with a practical, cost- and time-efficient method for VO2 estimation without the need for individual calibration.