二維運動估測測量股二頭肌位移及其在評估運動效果中的應用

Abstract

研究背景：運動估測是估計物體在不同時序之影像的運動過程，透過運動估測了解人體軟組織動態移動的特徵能協助擴展對於組織機械特性的認識，並有助於傷害機轉與復健運動成效的探討；超音波斑點追蹤方法即是一種應用超音波影像的運動估測方法，可以以此方法估計肌肉收縮時產生的位移與應變，來推論肌肉的收縮功能，然而目前於下肢肌肉的應用仍較少。股二頭肌長頭是大腿後肌最常見拉傷的肌肉，其可能受傷機制是離心收縮時產生過大的肌肉組織應變量而導致拉傷，而北歐大腿後肌彎舉運動是常被用作預防或治療大腿後肌拉傷的運動，然而目前並無研究探討在運動介入前後，肌肉組織於動態活動時位移量的變化。 研究目的：本研究包含兩部分，第一部分旨在應用超音波斑點追蹤方法於股二頭肌長頭肌肉組織上，以相似度比較方法歸一化交叉相關法(Normalized cross-correlation, NCC)所得的歸一化交叉相關係數NCC，結合再測信度，建立超音波斑點追蹤方法於股二頭肌長頭適切的施測方法學；第二部分將以八週北歐大腿後肌彎舉運動訓練介入健康受試者，並根據第一部分的超音波斑點追蹤施測方式來評估運動介入對肌肉組織位移量之效益。 研究方法：本研究第一部分預計納入10位健康男性受試者，以超音波斑點追蹤結合膝關節被動伸直、膝關節屈曲等長收縮、膝關節屈曲離心收縮等三項測試，在至少間隔一週的兩個不同時間點，對股二頭肌長頭肌肉組織進行運動估測，分別會安排量測在股二頭肌長頭肌肉近端及中段位置，以及受試者以最大和次大用力程度進行收縮，比較不同測試情境下的NCC值，和組織橫向位移量再測信度，同時也以剪力波彈性成像評估肌肉組織機械特性，嘗試解釋近端與中段結果具差異的可能原因；第二部分預計以八週北歐大腿後肌運動介入至少15位健康男性受試者，並以超音波斑點追蹤比較介入前後肌肉組織位移量的變化，同時也比較膝屈曲力矩、神經肌肉活化、肌肉形態、肌肉組織機械特性等參數之變化，以輔助解釋位移量的結果。 結果：在第一部分中，NCC結果顯示在股二頭肌肌肉中段量測位置結合最大收縮用力程度進行測試會得到較佳之NCC值，整體皆在0.80以上；再測信度結果則顯示有涉及關節活動的膝關節被動伸直（橫向位移量：ICC = 近端0.963 vs 中段0.406）、膝屈曲肌離心收縮測試（橫向位移量：ICC = 近端0.896 vs 中段0.600），在近端量測位置相較中段之信度較佳；在膝屈曲肌等長收縮測試則於中段量測位置相較近端為佳（橫向位移量：ICC = 近端0.738 vs 中段0.981）；雖被動伸直與離心收縮兩測試於近端量測位置信度較佳，其位移角度顯示肌肉組織於測試時是往坐骨方向移動，與理論及B模式影像上所見結果不同。第二部分實驗顯示慣用腳股二頭肌長頭肌肉組織在離心收縮時向遠端位移減少(Pre = 3.98 ± 3.84 mm vs Post = 1.50 ± 4.17 mm, P value = 0.006)，且皮爾森相關性分析結果顯示膝屈曲肌離心收縮力矩的進步和肌肉組織於離心收縮測試時向遠端位移量呈高度負相關(r = -0.63, P value < 0.01)，而在非慣用腳則並沒有以上之發現；另外發現介入後雙腳有顯著的膝屈曲力矩成長並伴隨雙腳肌束長度變長、肌肉厚度變厚、肌肉截面積增加等肌肥大適應，在神經肌肉活化方面僅慣用腳半腱肌肌電圖訊號尖峰值降低，其餘參數並無前後測或不同受測腳之顯著差異。 結論：第一部分結果顯示以股二頭肌長頭肌肉中段作為量測位置，以膝關節被動伸直、最大膝屈曲肌等長收縮、最大膝屈曲肌離心收縮測試來評估肌肉組織位移量是合適的量測方法，唯後續仍須提升再測信度；第二部分結果中，膝屈曲肌離心收縮力矩改善和肌肉組織向遠端位移量相關的結果，初步在人體驗證肌肉組織位移與大腿後肌拉傷受傷機制的關聯性，且肌肉組織在介入後較能抵抗離心收縮時向遠端延展的結果，解釋了北歐大腿後肌彎舉運動訓練降低拉傷風險的機制。在未來建議納入股二頭肌長頭不同區域肌肉以及肌腱組織做運動估測，且針對離心收縮時的肌束長度直接量測，將更有助於釐清傷害機制與運動介入效益。

Background: Motion estimation is the process of estimating the motion of objects in image in time series. It helps us to expand the understanding of tissue mechanical property, injury mechanism and effect of rehabilitation exercise by doing motion estimation on musculoskeletal soft tissue. Speckle-tracking ultrasonography is one of the motion estimation methods using ultrasound images. This method can be used to estimate the displacement and strain generated during muscle contraction to infer the dynamic function of muscle. However, the current application in lower limb muscles is still less. Biceps femoris long head (BFlh) is the most involved muscle in hamstring strain injury (HSI). Past studies supposed the injury mechanism was the excessive strain of muscle tissue during eccentric contraction. And the Nordic hamstring curl exercise (NHE) is often used to prevent or treat subjects with HSI. While there was no research on the changes in the displacement of muscular tissue during dynamic activities after exercise intervention. Purpose: This study consisted of two parts. The first part aimed to establish an appropriate methodology for speckle-tracking ultrasonography applying in the BFlh musculature by comparing the results of normalized cross-correlation, which is a similarity comparison method, and test-retest reliability. The second part aimed to evaluate the effect of exercise intervention on the displacement of muscular tissue according to the methodology suggested in the first part. Methods: Ten healthy male subjects were included in the first part of this study. We performed speckle-tracking ultrasonography scanning in three tests including passive knee extension test, knee flexion isometric contraction test and knee flexor eccentric contraction test at two time points at least one week apart. The measurement was conducted at the proximal and middle of the BFlh, and the subjects were tested in maximal and submaximal contraction intensity. The results of NCC and reliability of lateral displacement were compared to determine the appropriate methodology. Shear wave elastography was also used to assess muscular tissue mechanical property in an attempt to explain possible reasons for the difference in proximal and middle results. We planned to include fifteen healthy male subjects and assessed the effect of eight-week Nordic hamstring curl exercise intervention on the displacement of muscular tissue with speckle-tracking ultrasonography. At the same time, changes in parameters such as knee flexion torque, neuromuscular activation, muscular morphology, and mechanical property of muscular tissue were compared to assist explaining the results of displacement. Results: The first-part results showed that the measurement position in the middle of the biceps femoris muscle combined with the maximum contraction intensity would generate a better NCC value, which all were above 0.80. The test-retest reliability results showed that the passive knee extension and knee flexor eccentric contraction test demonstrated better outcome in muscle proximal part, which were ICC = 0.963 vs 0.406 in passive knee extension test and ICC = 0.896 vs 0.600 in knee flexor eccentric contraction test. In the knee flexion isometric contraction test, the measurement position in the middle was better than that in the proximal, which was ICC = 0.738 vs 0.981. Although the two tests of passive knee extension and knee flexor eccentric contraction showed more reliable in the proximal measurement position, the displacement angle results indicated that the muscular tissue moved toward the ischial tuberosity during the test which contradicting the results seen in theory and B-mode images. The second-part results demonstrated the BFlh muscle of dominant leg displaced less distally in knee flexor eccentric contraction test (Pre = 3.98 ± 3.84 mm vs Post = 1.50 ± 4.17 mm, P value = 0.006). Additionally, the results of Pearson correlation analysis showed negative correlation (r = -0.63, P value < 0.01) between the progress of knee flexor eccentric torque and the displacement of BFlh in knee flexor eccentric contraction test. However, the above findings were not found in the non-dominant leg. The results also showed bilateral knee flexion torque improved accompanied by muscle hypertrophy adaptations such as the fascicle length lengthened, muscle thickness thickened, and muscle cross-sectional area increased. In terms of neuromuscular activation, only peak amplitude of dominant semitendinosus muscle decreased. There was no significant difference in the other parameters before and after the tests or between different legs. Conclusion: The results in the first part of this study suggested that the middle part of the BFlh could be used as the measurement position. And the BFlh being evaluated by tests including passive knee extension test, maximal knee flexion isometric contraction and maximal knee flexor eccentric contraction was an appropriate methodology. However, the test-retest reliability still needed to be improved in the future. With the second-part results, we’ve verified the relationship between BFlh muscle tissue displacement and the injury mechanism of hamstring muscle strain in humans preliminarily. Also, we explained how the NHE training reduced the HSI injury risk with the results of muscle tissue being more resistant to stretch when the eccentric contraction. And the effect of NHE training might be more pronounced in the weaker leg. In the future, we recommended to include different part of BFlh muscle and tendon for motion estimation and measure the fascicle length when eccentric contraction. This would further help clarifying the injury mechanism and training effects on HSI.