Postural control in humans relies on information from receptors in the proprioceptive, visual, and vestibular systems of the body. As part of human aging, declines in all three postural control systems occur. Age-related changes impact multiple gait parameters, such as decreased range of motion in plantarflexion, increased hip flexion, and reduced stride length and gait velocity. In addition, excessive weight bearing on the heels during standing or forefoot-dominated walking creates risk factors for falls and injury within this population. Identification of these abnormal patterns by a computerized technique can help in early detection of gait changes and prevention of falls. This paper presents a case study to see if plantar pressure and heel acceleration patterns attributed to different motion activities can be accurately identified by a neural network classifier. The method has been tested on motion patterns collected from a single subject. The results show good sensitivity and specificity of the classifier, confirming the feasibility of further research.