Statistical humerus atlas for optimal design of Asian-specific humerus implants and associated intramedullary robotics

The statistical shape analysis of population-specific anatomical structures plays a pivotal role in both implant design and associated dedicated surgical robotic instruments. The sizes of current off-the-shelf humerus implants are unable to accommodate Asian patients since they are mainly produced for the American and European countries based on locally collected data. In this paper, an algorithm to construct a statistical humerus model is proposed so as to create a reference model for the development of humerus implants that conforms to the Asian population, and for the related design of snake-like robot for intramedullary procedures. The statistical humerus model is constructed from 14 CT datasets of male patients by implementing a three-step simplified method with improved alignment process. We have further constructed the statistical atlas showing the internal anatomy of humerus by adopting a spherical harmonics based method for the aim of developing intramedullary based implants used for humerus fixation. The validation results demonstrate that the proposed method significantly improves the efficiency of alignment without compromising the accuracy of the atlas. A comparison between the atlas constructed using the proposed method and that constructed utilizing the framework of spherical harmonics is further illustrated.