Campbell A C, Lloyd D G, Alderson J A, Elliott B C
School of Sport Science, Exercise and Health, The University of Western Australia, M408, 35 Stirling Highway, Crawley, WA 6009, Australia.
School of Sport Science, Exercise and Health, The University of Western Australia, M408, 35 Stirling Highway, Crawley, WA 6009, Australia.
J Biomech. 2009 Jul 22;42(10):1527-1532. doi: 10.1016/j.jbiomech.2009.03.039. Epub 2009 May 9.
Identification of the centre of the glenohumeral joint (GHJ) is essential for three-dimensional (3D) upper limb motion analysis. A number of convenient, yet un-validated methods are routinely used to estimate the GHJ location in preference to the International Society of Biomechanics (ISB) recommended methods. The current study developed a new regression model, and simple 3D offset method for GHJ location estimation, employing easy to administer measures, and compared the estimates with the known GHJ location measured with magnetic resonance imaging (MRI). The accuracy and reliability of the new regression and simple 3D offset techniques were compared with six established predictive methods. Twenty subjects wore a 3D motion analysis marker set that was also visible in MRI. Immediately following imaging, they underwent 3D motion analysis acquisition. The GHJ and anatomical landmark positions of 15 participants were used to determine the new regression and simple 3D generic offset methods. These were compared for accuracy with six established methods using 10 subject's data. A cross validation on 5 participants not used for regression model development was also performed. Finally, 10 participants underwent a further two MRI's and subsequent 3D motion analysis analyses for inter-tester and intra-tester reliability quantification. When compared with any of the other established methods, our newly developed regression model found an average GHJ location closer to the actual MRI location, having an GHJ location error of 13+/-2 mm, and had significantly lower inter-tester reliability error, 6+/-4 mm (p<0.01).
确定肩关节（GHJ）的中心对于三维（3D）上肢运动分析至关重要。许多方便但未经验证的方法通常被用于估计GHJ的位置，而不是国际生物力学学会（ISB）推荐的方法。本研究开发了一种新的回归模型和简单的3D偏移方法来估计GHJ的位置，采用易于实施的测量方法，并将估计值与通过磁共振成像（MRI）测量的已知GHJ位置进行比较。将新的回归和简单3D偏移技术的准确性和可靠性与六种既定的预测方法进行了比较。20名受试者佩戴了一个在MRI中也可见的3D运动分析标记集。成像后立即进行3D运动分析采集。15名参与者的GHJ和解剖标志位置用于确定新的回归和简单3D通用偏移方法。使用10名受试者的数据将这些方法与六种既定方法的准确性进行比较。还对5名未用于回归模型开发的参与者进行了交叉验证。最后，10名参与者又接受了两次MRI检查和随后的3D运动分析，以量化测试者间和测试者内的可靠性。与任何其他既定方法相比，我们新开发的回归模型发现平均GHJ位置更接近实际MRI位置，GHJ位置误差为13±2毫米，并且测试者间可靠性误差显著更低，为6±4毫米（p<0.01）。
