Linear and Nonlinear Modeling of Adult Malaysian Population´s Hand Grip Strength

Hand grip strength is a known indicator of hand function and its measurement is a useful criterion in designing post hand-surgery rehabilitation program. It is established that normal grip strength of patients is correlated with age, weight, height, gender, occupation and hand dominant. In this work, predictive linear and nonlinear models based on 412 healthy average Malaysians are built using least squares and neuro-fuzzy estimators respectively. Two quality of fits are used to evaluate the accuracy of the predictive models. It was found that the neuro-fuzzy model that maps the six inputs nonlinearly is superior to the linear model in describing the input-output relationship of the data