A real-time image analysis system for computer-assisted diagnosis of neurological disorders

This paper describes the application of digital image processing and patter n recognition techniques to assist in diagnosing neurological disorders. In medical practices the posture and movement of a human subject through his/ her gait cycle contains information that is used by an experienced clinicia n to determine the mental health of a patient. An image processing based sy stem can be used to automate this process and produce quantified results. T his is achieved by processing, extracting and classifying joint angle infor mation from still images of a human subject's gait. Several new algorithms are devised to process and extract the required information from the images : a new edge extraction technique is used to assist in image segmentation; a technique based on correlating images from different parts of a gait is u sed to obtain complete information about a subject's posture; and an algori thm based on the Hough Transform is used to obtain the limb joint angles an d swing distances of a subject. Joint angles and swing distances obtained f rom normal and patient subjects are then used in training and verifying cla ssifications using a feed-forward neural network and a fuzzy clustering alg orithm. (C) 1999 Academic Press.