Karine Fortin

Validity of the Microsoft Kinect for assessment of postural control

Clinically feasible methods of assessing postural control such as timed standing balance and functional reach tests provide important information, however, they cannot accurately quantify specific postural control mechanisms. The Microsoft Kinect™ system provides real-time anatomical landmark position data in three dimensions (3D), and given that it is inexpensive, portable and simple to setup it may bridge this gap. This study assessed the concurrent validity of the Microsoft Kinect™ against a benchmark reference, a multiple-camera 3D motion analysis system, in 20 healthy subjects during three postural control tests: (i) forward reach, (ii) lateral reach, and (iii) single-leg eyes-closed standing balance. For the reach tests, the outcome measures consisted of distance reached and trunk flexion angle in the sagittal (forward reach) and coronal (lateral reach) planes. For the standing balance test the range and deviation of movement in the anatomical landmark positions for the sternum, pelvis, knee and ankle and the lateral and anterior trunk flexion angle were assessed. The Microsoft Kinect™ and 3D motion analysis systems had comparable inter-trial reliability (ICC difference=0.06±0.05; range, 0.00-0.16) and excellent concurrent validity, with Pearsons r-values >0.90 for the majority of measurements (r=0.96±0.04; range, 0.84-0.99). However, ordinary least products analyses demonstrated proportional biases for some outcome measures associated with the pelvis and sternum. These findings suggest that the Microsoft Kinect™ can validly assess kinematic strategies of postural control. Given the potential benefits it could therefore become a useful tool for assessing postural control in the clinical setting.

Tibiofemoral contact forces during walking, running and sidestepping

We explored the tibiofemoral contact forces and the relative contributions of muscles and external loads to those contact forces during various gait tasks. Second, we assessed the relationships between external gait measures and contact forces. A calibrated electromyography-driven neuromusculoskeletal model estimated the tibiofemoral contact forces during walking (1.44±0.22ms(-1)), running (4.38±0.42ms(-1)) and sidestepping (3.58±0.50ms(-1)) in healthy adults (n=60, 27.3±5.4years, 1.75±0.11m, and 69.8±14.0kg). Contact forces increased from walking (∼1-2.8 BW) to running (∼3-8 BW), sidestepping had largest maximum total (8.47±1.57 BW) and lateral contact forces (4.3±1.05 BW), while running had largest maximum medial contact forces (5.1±0.95 BW). Relative muscle contributions increased across gait tasks (up to 80-90% of medial contact forces), and peaked during running for lateral contact forces (∼90%). Knee adduction moment (KAM) had weak relationships with tibiofemoral contact forces (all R(2)<0.36) and the relationships were gait task-specific. Step-wise regression of multiple external gait measures strengthened relationships (0.20<Radj(2)<0.78), but were variable across gait tasks. Step-wise regression equations from a particular gait task (e.g. walking) produced large errors when applied to a different gait task (e.g. running or sidestepping). Muscles well stabilized the knee, increasing their role in stabilization from walking to running to sidestepping. KAM was a poor predictor of medial contact force and load distributions. Step-wise regression models results suggest the relationships between external gait measures and contact forces cannot be generalized across tasks. Neuromusculoskeletal modelling may be required to examine tibiofemoral contact forces and role of muscle in knee stabilization across gait tasks.

Tibiofemoral Contact Forces in the Anterior Cruciate Ligament-Reconstructed Knee.

PURPOSE To investigate differences in anterior cruciate ligament-reconstructed (ACLR) and healthy individuals in terms of the magnitude of the tibiofemoral contact forces, as well as the relative muscle and external load contributions to those contact forces, during walking, running, and sidestepping gait tasks. METHODS A computational EMG-driven neuromusculoskeletal model was used to estimate the muscle and tibiofemoral contact forces in those with single-bundle combined semitendinosus and gracilis tendon autograft ACLR (n = 104, 29.7 ± 6.5 yr, 78.1 ± 14.4 kg) and healthy controls (n = 60, 27.5 ± 5.4 yr, 67.8 ± 14.0 kg) during walking (1.4 ± 0.2 m·s), running (4.5 ± 0.5 m·s) and sidestepping (3.7 ± 0.6 m·s). Within the computational model, the semitendinosus of ACLR participants was adjusted to account for literature reported strength deficits and morphological changes subsequent to autograft harvesting. RESULTS ACLR had smaller maximum total and medial tibiofemoral contact forces (~80% of control values, scaled to bodyweight) during the different gait tasks. Compared with controls, ACLR were found to have a smaller maximum knee flexion moment, which explained the smaller tibiofemoral contact forces. Similarly, compared with controls, ACLR had both a smaller maximum knee flexion angle and knee flexion excursion during running and sidestepping, which may have concentrated the articular contact forces to smaller areas within the tibiofemoral joint. Mean relative muscle and external load contributions to the tibiofemoral contact forces were not significantly different between ACLR and controls. CONCLUSIONS ACLR had lower bodyweight-scaled tibiofemoral contact forces during walking, running, and sidestepping, likely due to lower knee flexion moments and straighter knee during the different gait tasks. The relative contributions of muscles and external loads to the contact forces were equivalent between groups.

Diminished sub-maximal quadriceps force control in anterior cruciate ligament reconstructed patients is related to quadriceps and hamstring muscle dyskinesia

The aim of this study was to determine the effects of anterior cruciate ligament reconstruction (ACLR) on sub-maximal quadriceps force control with respect to quadriceps and hamstring muscle activity. Thirty ACLR individuals together with 30 healthy individuals participated. With real-time visual feedback of muscle force output and electromyographic electrodes attached to the quadriceps and hamstring muscles, subjects performed an isometric knee extension task where they increased and decreased their muscle force output at 0.128Hz within a range of 5-30% maximum voluntary capacity. The ACLR group completed the task with more error and increased medial hamstring and vastus medialis activation (p<0.05). Moderate negative correlations (p<0.05) were observed between quadriceps force control and medial (Spearmans rho=-0.448, p=0.022) and lateral (Spearmans rho=-0.401, p=0.034) hamstring activation in the ACLR group. Diminished quadriceps sub-maximal force control in ACLR subjects was reflective of medial quadriceps and hamstring dyskinesia (i.e., altered muscle activity patterns and coordination deficits). Within the ACLR group however, augmented hamstring co-activation was associated with better quadriceps force control. Future studies should explore the convergent validity of quadriceps force control in ACLR patients.

Lumbar extensor muscle force control is associated with disability in people with chronic low back pain

Background The ability to control lumbar extensor force output is necessary for daily activities. However, it is unknown whether this ability is impaired in chronic low back pain patients. Similarly, it is unknown whether lumbar extensor force control is related to the disability levels of chronic low back pain patients. Methods Thirty‐three chronic low back pain and 20 healthy people performed lumbar extension force‐matching task where they increased and decreased their force output to match a variable target force within 20%–50% maximal voluntary isometric contraction. Force control was quantified as the root‐mean‐square‐error between participants’ force output and target force across the entire, during the increasing and decreasing portions of the force curve. Within‐ and between‐group differences in force‐matching error and the relationship between back pain groups force‐matching results and their Oswestry Disability Index scores were assessed using ANCOVA and linear regression respectively. Findings Back pain group demonstrated more overall force‐matching error (mean difference = 1.60 [0.78, 2.43], P < 0.01) and more force‐matching error while increasing force output (mean difference = 2.19 [1.01, 3.37], P < 0.01) than control group. The back pain group demonstrated more force‐matching error while increasing than decreasing force output (mean difference = 1.74, P < 0.001, 95%CI [0.87, 2.61]). A unit increase in force‐matching error while decreasing force output is associated with a 47% increase in Oswestry score in back pain group (R2 = 0.19, P = 0.006). Interpretation Lumbar extensor muscle force control is compromised in chronic low back pain patients. Force‐matching error predicts disability, confirming the validity of our force control protocol for chronic low back pain patients. HighlightsLumbar extensor muscle force control is impaired in people with chronic low back pain.The ability to accurately increase, but not decrease, force output is impaired in people with chronic low back pain.Impairment in lumbar extensor muscle force control is associated with disability in people with chronic low back pain.

Accuracy of a novel marker tracking approach based on the low-cost Microsoft Kinect v2 sensor

Microsoft Kinect for Windows v2 is a motion analysis system that features a markerless human pose estimation algorithm. Given its affordability and portability, Kinect v2 has potential for use in biomechanical research and within clinical settings; however, recent studies suggest high inaccuracy of the markerless algorithm compared to marker-based motion capture systems. A novel tracking method was developed using Kinect v2, employing custom-made colored markers and computer vision techniques. The aim of this study was to test the accuracy of this approach relative to a conventional Vicon motion analysis system, performing a Bland-Altman analysis of agreement. Twenty participants were recruited, and markers placed on bony prominences near hip, knee and ankle. Three-dimensional coordinates of the markers were recorded during treadmill walking and running. The limits of agreement (LOA) of marker coordinates were narrower than - 10 and 10 mm in most conditions, however a negative relationship between accuracy and treadmill speed was observed along Kinect depth direction. LOA of the surrogate knee angles were within - 1.8°, 1.7° for flexion in all conditions and - 2.9°, 1.7° for adduction during fast walking. The proposed methodology exhibited good agreement with a marker-based system over a range of gait speeds and, for this reason, may be useful as low-cost motion analysis tool for selected biomechanical applications.

Cartilage quantitative T2 relaxation time 2-4 years following isolated anterior cruciate ligament reconstruction: CARTILAGE T2 FOLLOWING ISOLATED ACLR

Cartilage T2 relaxation time in isolated anterior cruciate ligament reconstruction (ACLR) without concomitant meniscal pathology and their changes over time remain unclear. The purpose of this exploratory study was to: (i) compare cartilage T2 relaxation time (T2 values) in people with isolated ACLR at 2–3 years post‐surgery (baseline) and matched healthy controls and; (ii) evaluate the subsequent 2‐year change in T2 values in people with ACLR. Twenty‐eight participants with isolated ACLR and nine healthy volunteers underwent knee magnetic resonance imaging (MRI) at baseline; 16 ACLR participants were re‐imaged 2 years later. Cartilage T2 values in full thickness, superficial layers, and deep layers were quantified in the tibia, femur, trochlear, and patella. Between‐group comparisons at baseline were performed using analysis of covariance adjusting for age, sex, and body mass index. Changes over time in the ACLR group were evaluated using paired sample t‐tests. ACLR participants showed significantly higher (p = 0.03) T2 values in the deep layer of medial femoral condyle at baseline compared to controls (mean difference 4.4 ms [13%], 95%CI 0.4, 8.3 ms). Over 2 years, ACLR participants showed a significant reduction (p = 0.04) in T2 value in the deep layer of lateral tibia (mean change 1.4 ms [−7%], 95%CI 0.04, 2.8 ms). The decrease in T2 values suggests improvement in cartilage composition in the lateral tibia (deep layer) of ACLR participants. Further research with larger ACLR cohorts divided according to meniscal status and matched healthy cohorts are needed to further understand cartilage changes post‐ACLR.

Individual muscle contributions to tibiofemoral compressive articular loading during walking, running and sidestepping

The tibiofemoral joint (TFJ) experiences large compressive articular contact loads during activities of daily living, caused by inertial, ligamentous, capsular, and most significantly musculotendon loads. Comparisons of relative contributions of individual muscles to TFJ contact loading between walking and sporting movements have not been previously examined. The purpose of this study was to determine relative contributions of individual lower-limb muscles to compressive articular loading of the medial and lateral TFJ during walking, running, and sidestepping. The medial and lateral compartments of the TFJ were loaded by a combination of medial and lateral muscles. During all gait tasks, the primary muscles loading the medial and lateral TFJ were the vastus medialis (VM) and vastus lateralis (VL) respectively during weight acceptance, while typically the medial gastrocnemii (MG) and lateral gastrocnemii (LG) dominated medial and lateral TFJ loading respectively during midstance and push off. Generally, the contribution of the quadriceps muscles were higher in running compared to walking, whereas gastrocnemii contributions were higher in walking compared to running. When comparing running and sidestepping, contributions to medial TFJ contact loading were generally higher during sidestepping while contributions to lateral TFJ contact loading were generally lower. These results suggests that after orthopaedic procedures, the VM, VL, MG and LG should be of particular rehabilitation focus to restore TFJ stability during dynamic gait tasks.

Trunk and lower limb coordination during lifting in people with and without chronic low back pain

Differences in synchronous movement between the trunk and lower limb during lifting have been reported in chronic low back pain (CLBP) patients compared to healthy people. However, the relationship between movement coordination and disability in CLBP patients has not been investigated. A cross-sectional study was conducted to compare regional lumbar and lower limb coordination between CLBP (n = 43) and control (n = 29) groups. The CLBP group was divided into high- and low-disability groups based on their Oswestry Disability Index (ODI) score. The mean absolute relative phase (MARP) angles and mean deviation phase (DP) between the (1) lumbar spine and hip, and (2) hip and knee were measured. The relationship between MARP angle and DP and ODI were investigated using linear regression. The higher-disability CLBP group demonstrated significantly greater lumbar-hip MARP angles than the lower-disability CLBP group (mean difference = 12.97, % difference = 36, p = 0.041, 95% CI [2.97, 22.98]). The higher-disability CLBP group demonstrated significantly smaller hip-knee DP than controls (mean difference = 0.11, % difference = 76, p = 0.011, 95% CI [0.03, 0.19]). There were no significant differences in lumbar-hip and hip-knee MARP and DP between the lower-disability CLBP and control groups. Lumbar-hip MARP was positively associated with ODI (R2 = 0.092, β = 0.30, p = 0.048). High-disability CLBP patients demonstrated decreased lumbar-hip movement coordination and stiffer hip-knee movement during lifting than low-disability CLBP patients and healthy controls.

Relationships Between Tibiofemoral Contact Forces and Cartilage Morphology at 2 to 3 Years After Single-Bundle Hamstring Anterior Cruciate Ligament Reconstruction and in Healthy Knees

Background: Prevention of knee osteoarthritis (OA) following anterior cruciate ligament (ACL) rupture and reconstruction is vital. Risk of postreconstruction knee OA is markedly increased by concurrent meniscal injury. It is unclear whether reconstruction results in normal relationships between tibiofemoral contact forces and cartilage morphology and whether meniscal injury modulates these relationships. Hypotheses: Since patients with isolated reconstructions (ie, without meniscal injury) are at lower risk for knee OA, we predicted that relationships between tibiofemoral contact forces and cartilage morphology would be similar to those of normal, healthy knees 2 to 3 years postreconstruction. In knees with meniscal injuries, these relationships would be similar to those reported in patients with knee OA, reflecting early degenerative changes. Study Design: Cross-sectional study; Level of evidence, 3. Methods: Three groups were examined: (1) 62 patients who received single-bundle hamstring reconstruction with an intact, uninjured meniscus (mean age, 29.8 ± 6.4 years; mean weight, 74.9 ± 13.3 kg); (2) 38 patients with similar reconstruction with additional meniscal injury (ie, tear, repair) or partial resection (mean age, 30.6 ± 6.6 years; mean weight, 83.3 ± 14.3 kg); and (3) 30 ligament-normal, healthy individuals (mean age, 28.3 ± 5.2 years; mean weight, 74.9 ± 14.9 kg) serving as controls. All patients underwent magnetic resonance imaging to measure the medial and lateral tibial articular cartilage morphology (volumes and thicknesses). An electromyography-driven neuromusculoskeletal model determined medial and lateral tibiofemoral contact forces during walking. General linear models were used to assess relationships between tibiofemoral contact forces and cartilage morphology. Results: In control knees, cartilage was thicker compared with that of isolated and meniscal-injured ACL-reconstructed knees, while greater contact forces were related to both greater tibial cartilage volumes (medial: R 2 = 0.43, β = 0.62, P = .000; lateral: R 2 = 0.19, β = 0.46, P = .03) and medial thicknesses (R 2 = 0.24, β = 0.48, P = .01). In the overall group of ACL-reconstructed knees, greater contact forces were related to greater lateral cartilage volumes (R 2 = 0.08, β = 0.28, P = .01). In ACL-reconstructed knees with lateral meniscal injury, greater lateral contact forces were related to greater lateral cartilage volumes (R 2 = 0.41, β = 0.64, P = .001) and thicknesses (R 2 = 0.20, β = 0.46, P = .04). Conclusion: At 2 to 3 years postsurgery, ACL-reconstructed knees had thinner cartilage compared with healthy knees, and there were no positive relationships between medial contact forces and cartilage morphology. In lateral meniscal-injured reconstructed knees, greater contact forces were related to greater lateral cartilage volumes and thicknesses, although it was unclear whether this was an adaptive response or associated with degeneration. Future clinical studies may seek to establish whether cartilage morphology can be modified through rehabilitation programs targeting contact forces directly in addition to the current rehabilitation foci of restoring passive and dynamic knee range of motion, knee strength, and functional performance.