A statistical approach to data analysis and 3-D geometric description of the human head and face.

Many analytical biomechanical methods require extensive three-dimensional statistical description of anatomical geometry. In particular, to design personal protective items for the human head and face, where good fit is critical, it is inevitable that a three-dimensional statistical description of this complicated structure will be needed. The work here offers an approach to this problem. This approach consists of three steps: (1) osteometric scaling, (2) normative specimen accumulation and (3) statistical testing. Three groups of facial data (24 Asian, 29 Black, and 29 White) were digitized. The effectiveness and accuracy of the statistical approach was tested on these three different experimental specimen sets. The method was found to be very accurate in modelling the most complicated human body parts--head and face. The availability of this detailed geometric information will also open many doors for future research and development of muscle controlled prostheses, repair of ligament damage, and in-vivo bone remodelling.