Automatic reconstruction of the muscle architecture from the superficial layer fibres data

Abstract

Background and objective: Physiological cross-sectional area (PCSA) of a muscle plays a significant role in determining the force contribution of muscle fascicles to skeletal movement. This parameter is typically calculated from the lengths of muscle fibres selectively sampled from the superficial layer of the muscle. However, recent studies have found that the length of fibres in the superficial layer often differs significantly (p < 0.5) from the length of fibres in the deep layer. As a result, PCSA estimation is inaccurate. In this paper, we propose a method to automatically reconstruct fibres in the whole volume of a muscle from those selectively sampled on the superficial layer.Methods:The method performs a centripetal Catmull–Rom interpolation of the input fibres within the volume of a muscle represented by its 3D surface model, automatically distributing the fibres among multiple heads of the muscle and shortening the deep fibres to support large attachment areas with extremely acute angles.Results:Our C++ implementation runs in a couple of seconds on commodity hardware providing realistic results for both artificial and real data sets we tested.Conclusions:The fibres produced by the method can be used directly to determine the personalised mechanical muscle functioning. Our implementation is publicly available for the researchers at https://mi.kiv.zcu.cz/.