RESUMEN
This study sought to establish a set of hip prostheses for the Chinese population and improve the design guidelines for hip prostheses. 480 cadaveric adult femora were obtained for X-ray imaging. 21 geometric parameters of femora were sampled on the anteroposterior(AP) and mediolateral(ML) views of radiograph in conjunction with computer-aided image processing technology. The curves of canal of proximal femur were fitted and made statistics. Pattern recognition was used to classify these geometric parameters. The results of classification corresponded with the 10 groups of data for design. On the basis of the numerous set of data, the results and classification would serve as means and guidelines for optimizing the design of prosthetic components and so far as custom-built hip prosthetics.