A Research of Bone Maturation Evaluation System Based on Active Shape Models
HU Yong-li, WANG Wei, YIN Bao-cai (Beijing Municipal Key Laboratory Multimedia and Intelligent Software Technology Beijing Polytechnic University, Beijing 100022)
A bone maturation evaluation system based on Active Shape Models is reported in this paper. By the system the bones in a hand X-radiographic image are located and their edges are detected. According to the standard of Chinese wrist bone maturation--CHN, the bones are sorted into different grades. Having gotten all grades of the bones in the hand image, the data of bone maturation can be computed by combination of the grades with different weights. The edges of the bones in X-radiograph image are faint and the grades' definitions of CHN are very abstractive. It is difficult to detect the edges of the bones and make correct classification of them. The Active Shape Models, based on the prior knowledge of the bones' edges, is used to detect the edges of the bones and vectorize the grades' definitions in CHN. The experiments reveal that the method of Active Shape Models is valid in bones' edges detection and grades classification. Tested with a set of hand X-radiograph images, the bone maturation evaluation system has good accuracy and stability.