Cylindrical depth image based customized helical bone plate design.

Commercial straight metal plates have been generally used to fix fractured bones, but recently, the need for customized and helical metal plates has emerged. Customized metal plates are designed to fit the shape of the fracture area that is a 3D curved surface, making it more difficult than designing on a 2D plane. Helical plates are researched due to their advantage in avoiding blood vessel damage compared to commercially available straight metal plates. In this paper, we propose a novel algorithm to design a customized helical metal plate for the femur using cylindrical depth images and Boolean operations. We also present the results of 3D printing a metal plate designed using the proposed algorithm, and the shape matching is verified by calculating the minimum distance between the surface of the printed plate and the surface of the femur.

An Algorithm for Computing Side Chain Conformational Variations of a Protein Tunnel/Channel.

In this paper, a novel method to compute side chain conformational variations for a protein molecule tunnel (or channel) is proposed. From the conformational variations, we compute the flexibly deformed shapes of the initial tunnel, and present a way to compute the maximum size of the ligand that can pass through the deformed tunnel. By using the two types of graphs corresponding to amino acids and their side chain rotamers, the suggested algorithm classifies amino acids and rotamers which possibly have collisions. Based on the divide and conquer technique, local side chain conformations are computed first, and then a global conformation is generated by combining them. With the exception of certain cases, experimental results show that the algorithm finds up to 327,680 valid side chain conformations from 128~1233 conformation candidates within three seconds.