Document Type



Accurate prediction of the lumbosacral and hip joint centres (LSJC, HJC) is crucial for the analysis of lumbo-pelvic-hip dynamics in various movements. Here we show that pre-existing regressions proposed by European research groups produce biased estimates of the LSJC and HJC in Japanese males and females (n = 23, 24), and that the biases in LSJC depend on sex. When compared to locations directly measured by MRI, the pre-existing regressions estimated the LSJC to be more posterior in males and more inferior and posterior in females, and the HJC to be more medial for both sexes. We suggest the importance of considering racial and sexual differences in morphology. We propose new regressions for Asians and validate them using leave-one-out cross-validation. Our regression can be a powerful solution for accurate motion analysis in Asians.