Investigating the Influence of Obesity on Gait Using Support Vector Machine Analysis

As the prevalence of obesity continues to increase at the population level, there is increasing concern about the impact of co-morbidities such as osteoarthritis. Osteoarthritis is a loading-related disease and its development and progression have been linked to excess body weight [1]. Changes in gait biomechanics, particularly at the knee, have been associated with increasing osteoarthritis severity in older adults. Differences in knee flexion excursion, peak knee flexion, peak knee adduction, peak external knee flexion moment, and peak external knee adduction moment have been reported [2,3]. However, it should be noted that these differences in knee variables with osteoarthritis were found in older adults. As the proportion of young people who are obese also rises, it is important to investigate the risk of osteoarthritis development at a younger age. This may be linked to patterns of knee biomechanics during gait. Simple statistical comparisons of discrete knee variables among young adults grouped by body mass index (BMI) found no differences among the groups in a previous report from our study [4]. Any relationship between knee biomechanics and osteoarthritis risk may be more complex in younger adults. Therefore, the purpose of this study was to further examine potential differences in knee biomechanics during gait among normal BMI, overweight, and obese young adults, using support vector machines [5], which are a group of pattern recognition and classification techniques.