FEATURE-EXTRACTION IN CARPAL-BONE ANALYSIS

Hand-bone analysis with image processing techniques using a digital ra diograph can be used to assess skeletal age. It consists of two steps, phalangeal and carpal bone analysis. This paper presents the carpal b one analysis. First, the carpal bone region of interest (CROI) is defi ned using a standard thresholding technique to separate the hand from the background. Then, a dynamic thresholding method with variable wind ow sizes is used to differentiate between the bones from the soft tiss ue. Next, the radius, ulna and metacarpals intersecting the borders of the CROI are removed by using mathematical morphology. Finally, all o bjects included in the corrected CROI are separated and described in t erms of features. These features describe the size, shape, and locatio n. They also include some gray scale pixel value information. On the b asis of this analysis the separation of the noncarpal bone objects fro m the carpal bones is possible. Feature selection step removes feature s of low discriminant power and reduces the space dimension. The remai ned carpal bone parameters are used for further analysis leading to th e skeletal age assessment.