Estimation of musculotendon parameters for scaled and subject specific musculoskeletal models using an optimization technique

详细信息    查看全文

• 作者：Luca Modenese^a ; ^b ; ^c ; ^{l.modenese@sheffield.ac.uk" class="auth_mail" title="E-mail the corresponding author} ; Elena Ceseracciu^d ; Monica Reggiani^d ; David G. Lloyd^a
• 关键词：Optimal fiber length ; Tendon slack length ; Scaling ; Subject specific ; Muscle models ; Hip joint ; Muscle&ndash ; tendon parameters ; Parameter optimization
• 刊名：Journal of Biomechanics
• 出版年：2016
• 出版时间：25 January 2016
• 年：2016
• 卷：49
• 期：2
• 页码：141-148
• 全文大小：2609 K

文摘

A challenging aspect of subject specific musculoskeletal modeling is the estimation of muscle parameters, especially optimal fiber length and tendon slack length. In this study, the method for scaling musculotendon parameters published by Winby et al. (2008), J. Biomech. 41, 1682–1688, has been reformulated, generalized and applied to two cases of practical interest: 1) the adjustment of muscle parameters in the entire lower limb following linear scaling of a generic model and 2) their estimation “from scratch” in a subject specific model of the hip joint created from medical images. In the first case, the procedure maintained the muscles׳ operating range between models with mean errors below 2.3% of the reference model normalized fiber length value. In the second case, a subject specific model of the hip joint was created using segmented bone geometries and muscle volumes publicly available for a cadaveric specimen from the Living Human Digital Library (LHDL). Estimated optimal fiber lengths were found to be consistent with those of a previously published dataset for all 27 considered muscle bundles except gracilis. However, computed tendon slack lengths differed from tendon lengths measured in the LHDL cadaver, suggesting that tendon slack length should be determined via optimization in subject-specific applications. Overall, the presented methodology could adjust the parameters of a scaled model and enabled the estimation of muscle parameters in newly created subject specific models. All data used in the analyses are of public domain and a tool implementing the algorithm is available at https://simtk.org/home/opt_muscle_par.