利用無標記運動捕捉系統評估退化性腰椎疾病患者下肢運動學特徵的信度與效度研究: 以坐到站測試為例

Abstract

背景： 退化性腰椎疾病(Degenerative lumbar disease, DLD)是一種常見的老年疾病，會導致腰椎退化以及下肢功能障礙，包含下肢疼痛、無力、感覺改變；下背疼痛、軀幹無力、神經性跛行、與平衡控制能力缺損。研究發現，這些患者在進行臨床五次坐站測試時其下肢運動學存在明顯的異常和改變。過去這些運動學變化主要是由標記式動作捕捉系統偵測，然而此系統的複雜設置卻不便於運用在臨床。隨著科技發展，無標記式動作捕捉系統由於其便攜性強、使用簡易、能檢測人體關節位置等優點，若臨床上的功能性測試輔以無標記動作捕捉系統對患者做評估，可得到的臨床資訊除了時間之參數，更可擴及空間之參數，實現直接反映患者三維動作特徵之優勢。 研究目的：驗證無標記式動作捕捉系統在檢測年輕健康人上，並應用至退化性腰椎疾病患者的下肢運動學表現，以提供新興評估工具其遠距醫療應用支持性證據。 研究假說：驗證無標記式動作捕捉系統於年輕健康人中預計有良好的信度與效度，並可應用於退化性腰椎狹窄的病患，進一步評估運動學相關資訊。 研究設計：橫斷型研究 研究方法：本研究共招募11位(年紀:27.28 ± 6.92歲)無其他骨骼肌肉方面疾病之年輕健康人，以及10位(年紀:70.00 ± 8.08歲)經國立臺灣大學附設醫院神經外科醫師診斷為退化性腰椎疾病患者，量測內容包含基本資料與下肢段長度量測，並同時使用標記式動作捕捉系統與無標記式動作捕捉系統，偵測受測者動作，以量化五次坐到站測試之結果。統計分析使用組內相關係數，描述標記式動作捕捉系統與無標記式動作捕捉系統於空間參數的可重複性；標記式動作捕捉系統與無標記式動作捕捉系統的絕對一致性使用皮爾森相關係數作為描述，顯著程度設定於0.01；兩組(年輕健康人與退化性腰椎疾病患者)運動學資料之比較使用獨立樣本t檢定分析，顯著程度設定於0.05。 結果：標記式動作捕捉系統與無標記式動作捕捉系統，在下肢髖關節、膝關節與踝關節的各軸向於健康人具有高的再測信度表現；且不論是在健康人或是退化性腰椎疾病患者其應用上皆具有高到非常高的共同效度表現 (p < 0.01)。在動作學資料應用上顯示退化性腰椎疾病患者比健康人需要花更長時間完成五次坐到站動作 (p < 0.05)；在動作期間關節移動範圍顯示退化性腰椎疾病患者比健康人有較小的關節活動 (p < 0.05)；此外，退化性腰椎疾病患者比健康人有更慢的關節移動角速度 (p < 0.05)。 結論：使用無標記動作評估系統應用在退化性腰椎疾病患者具有良好的信效度表現，且評估下肢動態學特徵時，可提供量化之運動學參數，並與健康人做出區別。可為臨床提供治療介入前後的運動學客觀參數，將助於近一步實踐遠距醫療評估的實施。

Background: Degenerative lumbar disease (DLD) is very common in older adults and leads to lumbar degeneration and lower extremity dysfunction. Previous studies have shown that these patients exhibit significant abnormalities and changes in lower extremity kinematics when performing the five times sit-to-stand test clinically. Previously, marker-based motion capture systems were used to detect pathological changes, which were not easy to use in clinics because of the complex settings. With advances in technology, markerless motion capture systems have become popular. They are easy to carry around, simple to use, and can track how our joints move without needing markers. Using markerless motion capture systems into functional performance tests for patient evaluation may provide temporal parameters and extend to spatiotemporal parameters, directly quantifying 3D kinematic data of the patient population. Purpose: Validation of markerless motion capture systems for assessing lower limb kinematics in young, healthy individuals and application to patients with DLD. This study aimed to establish the reliability and validity of using a markerless motion capture system and to examine the 3D kinematic changes during the five times sit-to-stand test in DLD patients. Design: Cross-sectional study Methods: This study recruited 11 young, healthy individuals (27.28 ± 6.92 years old) without other musculoskeletal disorders and 10 patients (70.00 ± 8.08 years old) diagnosed with degenerative lumbar disease by neurosurgeons at National Taiwan University Hospital. Measurements included basic demographic information and lower limb length measurements. Both groups were subjected to five times sit-to-stand tests using marker-based and markerless motion capture systems. Statistical analysis involved assessing the repeatability of spatial parameters between marker-based and markerless motion capture systems using the Intraclass Correlation Coefficient (ICC (3,k)). The correlation between the two systems was also evaluated using Pearson's correlation coefficient, with a significant level set at 0.01. The comparison of kinematic data between the two groups (young healthy individuals and patients with DLD) was analyzed using independent sample t-tests, with the significance level set at 0.05 Results: Both marker-based and markerless motion capture systems showed high test-retest reliability for the hip, knee, and ankle in healthy individuals. Additionally, both systems demonstrated high concurrent validity in applications for healthy individuals and patients with DLD (p < 0.01). Kinematic data analysis revealed that patients with DLD spent more time completing five sit-to-stand than healthy individuals (p < 0.05). During the movement, patients with DLD showed a smaller joint excursion to healthy individuals (p < 0.05). Additionally, patients with DLD exhibited slower joint angular velocities than healthy individuals (p < 0.05). Conclusions: It is anticipated that when using markerless motion capture systems to evaluate kinematic changes in lower extremities, the spatiotemporal parameters would demonstrate good performance. Conducting kinematic assessments in DLD patients using markerless motion capture systems provides objective kinematic parameters for clinical interventions and facilitates the practical implementation of telemedicine evaluations.