Pose2Sim: An End-to-End Workflow for 3D Markerless Sports Kinematics-Part 2: Accuracy.

Two-dimensional deep-learning pose estimation algorithms can suffer from biases in joint pose localizations, which are reflected in triangulated coordinates, and then in 3D joint angle estimation. Pose2Sim, our robust markerless kinematics workflow, comes with a physically consistent OpenSim skeletal model, meant to mitigate these errors. Its accuracy was concurrently validated against a reference marker-based method. Lower-limb joint angles were estimated over three tasks (walking, running, and cycling) performed multiple times by one participant. When averaged over all joint angles, the coefficient of multiple correlation (CMC) remained above 0.9 in the sagittal plane, except for the hip in running, which suffered from a systematic 15° offset (CMC = 0.65), and for the ankle in cycling, which was partially occluded (CMC = 0.75). When averaged over all joint angles and all degrees of freedom, mean errors were 3.0°, 4.1°, and 4.0°, in walking, running, and cycling, respectively; and range of motion errors were 2.7°, 2.3°, and 4.3°, respectively. Given the magnitude of error traditionally reported in joint angles computed from a marker-based optoelectronic system, Pose2Sim is deemed accurate enough for the analysis of lower-body kinematics in walking, cycling, and running.

Pose2Sim: An End-to-End Workflow for 3D Markerless Sports Kinematics-Part 1: Robustness.

Being able to capture relevant information about elite athletes' movement "in the wild" is challenging, especially because reference marker-based approaches hinder natural movement and are highly sensitive to environmental conditions. We propose Pose2Sim, a markerless kinematics workflow that uses OpenPose 2D pose detections from multiple views as inputs, identifies the person of interest, robustly triangulates joint coordinates from calibrated cameras, and feeds those to a 3D inverse kinematic full-body OpenSim model in order to compute biomechanically congruent joint angles. We assessed the robustness of this workflow when facing simulated challenging conditions: (Im) degrades image quality (11-pixel Gaussian blur and 0.5 gamma compression); (4c) uses few cameras (4 vs. 8); and (Cal) introduces calibration errors (1 cm vs. perfect calibration). Three physical activities were investigated: walking, running, and cycling. When averaged over all joint angles, stride-to-stride standard deviations lay between 1.7° and 3.2° for all conditions and tasks, and mean absolute errors (compared to the reference condition-Ref) ranged between 0.35° and 1.6°. For walking, errors in the sagittal plane were: 1.5°, 0.90°, 0.19° for (Im), (4c), and (Cal), respectively. In conclusion, Pose2Sim provides a simple and robust markerless kinematics analysis from a network of calibrated cameras.

ILIOTIBIAL BAND SYNDROME IN CYCLING: A COMBINED EXPERIMENTAL-SIMULATION APPROACH FOR ASSESSING THE EFFECT OF SADDLE SETBACK.

BACKGROUND: Despite abundant literature, the treatment of iliotibial band syndrome (ITBS) in cyclists remains complicated as it lacks evidence-based recommendations. PURPOSE: The aim of this study was to develop a musculoskeletal modelling approach that investigates three potential biomechanical determinants of ITBS (strain, strain rate and compression force) and to use this approach to investigate the effect of saddle setback. DESIGN: Cross-sectional. METHODS: An existing 3D lower-body musculoskeletal model was adapted to cycling and to the computation of three putative pathomechanisms responsible for ITBS: ITB strain, ITB strain rate, and compression force between ITB and the lateral femoral epicondyle (LFE). Lower limb kinematics recorded from ten well-trained healthy cyclists served as input data of the model. Cyclists pedalled at a steady state (90rpm and 200W) on an ergometer, and three different saddle setback conditions were tested. The theoretical combined influence of hip and knee joint angles on ITBS was investigated and analysed through the lens of individual pedalling technique. RESULTS: ITB-LFE compression force was the only parameter significantly affected by saddle setback and supports the hypothesis that compression force is likely to be a determinant factor in ITBS etiology. Furthermore, results showed that ITB-LFE compression force increases in individuals whose pedalling technique exacerbates hip extension-adduction and/or knee extension-internal rotation. CONCLUSION: This approach has the potential to be advantageously implemented as an additional tool to help diagnose/correct potentially harmful sport techniques and optimize equipment setup/design. LEVEL OF EVIDENCE: 3b.

Influence of saddle setback on knee joint forces in cycling.

Knee functional disorders are one of the most common lower extremity non-traumatic injuries reported by cyclists. Incorrect bicycle configuration may predispose cyclist to injury but the evidence of an effect of saddle setback on knee pain remains inconclusive. The aim of this study was to determine the effect of saddle setback on knee joint forces during pedalling using a musculoskeletal modelling approach. Ten cyclists were assessed under three saddle setback conditions (range of changes in saddle position ~6 cm) while pedalling at a steady power output of 200 W and cadence of 90 rpm. A cycling musculoskeletal model was developed and knee joint forces were estimated using an inverse dynamics method associated with a static optimisation procedure. Our results indicate that moving the saddle forwards was not associated with an increase of patellofemoral joint forces. On the contrary, the tibiofemoral mean and peak compression force were 14 and 15% higher in the Backward than in the Forward condition, respectively. The peak compression force was related to neither pedal force nor quadriceps muscle force but coincided with the eccentric contraction of knee flexor muscles. These findings should benefit bike fitting practitioners and coaches in the design of specific training/rehabilitation protocols.

Influence of saddle setback on pedalling technique effectiveness in cycling.

Besides its regulation by Union Cycliste Internationale, the evidence relating saddle setback to pedalling performance remains inconclusive. This study investigates the influence of saddle setback on pedalling effectiveness through two indexes: an index of pedalling force effectiveness and an index of pedalling work effectiveness. Eleven cyclists were assessed six saddle setback conditions while pedalling at a steady power output of 200 W and cadence of 90 rpm. A force sensor was integrated within the seat post to compute the centre of pressure on the saddle. From instrumented pedals, an index of force effectiveness (ratio between the force directed perpendicular to the crank arm and the total force applied to the pedal) and an index of work effectiveness (based on the minimisation of negative crank work) were calculated. In comparison with a forward position, sitting backward significantly decreased 5% cumulative total work, increased index of work effectiveness (84.2 ± 3.7 vs. 82.0 ± 4.7%), and increased index of force effectiveness (41.7 ± 2.9 vs. 39.9 ± 3.7 and 36.9 ± 0.7%). Thus, while it was previously reported that sitting more forward favours maximal power, this study demonstrates that it also leads to a decreased effectiveness in steady-state pedalling.

Influence of nerve supply on hand electromyography coherence during a three-digit task.

Intermuscular coupling has been investigated to understand neural inputs to coordinate muscles in a motor performance. However, little is known on the role of nerve innervation on intermuscular coupling. The purpose of this study was to investigate how the anatomy of nerve distribution affected intermuscular coupling in the hand during static grip. Electromyographic (EMG) signals were recorded from intrinsic and extrinsic muscles while subjects performed a static grip. Coherence was computed for muscle pairs innervated by either the same or different nerves. The results did not support the hypothesis that muscles sharing the same nerve exhibit greater coupling than muscles innervated by different nerves. In general, extrinsic muscle pairs displayed higher coherence than intrinsic pairs. The results suggest that intermuscular coupling in a voluntary motor task is likely modulated in a functional manner and that different nerves might transport common neural inputs to functionally coupled muscles.

Movement of the distal carpal row during narrowing and widening of the carpal arch width.

Change in carpal arch width (CAW) is associated with wrist movement, carpal tunnel release, or therapeutic tunnel manipulation. This study investigated the angular rotations of the distal carpal joints as the CAW was adjusted. The CAW was narrowed and widened by 2 and 4 mm in seven cadaveric specimens while the bone positions were tracked by a marker-based motion capture system. The joints mainly pronated during CAW narrowing and supinated during widening. Ranges of motion about the pronation axis for the hamate-capitate (H-C), capitate-trapezoid (C-Td), and trapezoid-trapezium (Td-Tm) joints were 8.1 ± 2.3 deg, 5.3 ± 1.3 deg, and 5.5 ± 3.5 deg, respectively. Differences between the angular rotations of the joints were found at ΔCAW = -4 mm about the pronation and ulnar-deviation axes. For the pronation axis, angular rotations of the H-C joint were larger than that of the C-Td and Td-Tm joints. Statistical interactions among the factors of joint, rotation axis, and ΔCAW indicated complex joint motion patterns. The complex three-dimensional motion of the bones can be attributed to several anatomical constraints such as bone arrangement, ligament attachments, and articular congruence. The results of this study provide insight into the mechanisms of carpal tunnel adaptations in response to biomechanical alterations of the structural components.

A digit alignment device for kinematic analysis of the thumb and index finger.

Kinematic analysis of the digits using optical motion capture systems relies on defining accurate coordinate systems for the individual segments. Limitations of previous digit kinematic protocols include marker placement errors, marker occlusion and superimposition, and skin movement artifact. The purpose of this study was to develop a protocol utilizing a digit alignment device (DAD) and nail marker clusters to overcome these limitations. Ten subjects underwent 10 static calibration trials for validation. The orientation of the thumb distal phalange relative to the index finger distal phalange was described using Euler angles of pitch(x), yaw(y'), and roll(z''). The digit calibration protocol demonstrated high accuracy (0.5°, 1.9° and 2.2° for x, y', z'') and precision (1.4°, 2.3° and 3.1° for x, y', z''). The developed protocol provided convenient identification of transformations that determine anatomically relevant coordinate systems for the distal phalanges of the digits. The potential of utilizing this protocol as a standardized tool for digit kinematics was demonstrated using a dynamic task of precision pinching.

Influence of index finger proximal interphalangeal joint arthrodesis on precision pinch kinematics.

PURPOSE: To evaluate the impact of proximal interphalangeal (PIP) joint arthrodesis on the kinematics of precision pinch. METHODS: Eleven healthy subjects performed index finger-thumb pinch motions under 4 conditions: unrestricted thumb and index finger (CONTROL) and fusion of the PIP joint of the index finger in flexion of 30° (PIP30), 40° (PIP40), and 50° (PIP50). Fusion was simulated with metallic splints. Kinematics of the thumb and index finger were recorded with a motion capture system. RESULTS: Proximal interphalangeal joint fusion at 30°, 40°, and 50° restricted maximal pinch span between the thumb tip and index finger tip by 6%, 10%, and 14%, respectively. At the time of pulp contact, PIP fusion led to an increase in index metacarpophalangeal joint flexion angle for the PIP30 condition and an increase in variability of thumb tip location for the PIP50 condition. Furthermore, the dynamic coordination between joint angles throughout the movement was affected by PIP fusion. CONCLUSIONS: This study reports impairment in the kinematics of precision pinch associated with index finger PIP joint fusion. A PIP joint fusion at 40° to 50° leads to a more natural precision pinch posture, but it restricts the aperture and reduces pinch precision.

Effect of object width on muscle and joint forces during thumb-index finger grasping.

The objective of this study was to identify the impact of modifying the object width on muscle and joint forces while gripping objects. The experimental protocol consisted to maintain horizontally five objects of different widths (3.5, 4.5, 5.5, 6.5, and 7.5 cm) with a thumb-index finger grip. Subjects were required to grasp spontaneously the object without any instruction regarding the grip force (GF) to apply. A biomechanical model of thumb-index finger pinch was developed to estimate muscle and joint forces. This model included electromyography, fingertip force, and kinematics data as inputs. The finger joint postures and the GF varied across the object widths. The estimated muscle forces also varied significantly according to the object width. Interestingly, we observed that the muscle force/GF ratios of major flexor muscles remain particularly stable with respect to the width whereas other muscle ratios differed largely. This may argue for a control strategy in which the actions of flexors were preserved in spite of change in joint postures. The estimated joint forces tended to increase with object width and increased in the distal-proximal sense. Overall, these results are of importance for the ergonomic design of handheld objects and for clinical applications.

Biomechanical effect of increasing or decreasing degrees of freedom for surgery of trapeziometacarpal joint arthritis: a simulation study.

Osteoarthritis of the trapeziometacarpal (TMC) joint can be treated by arthrodesis and arthroplasty, which potentially decreases or increases the degrees of freedom (DoF) of the joint, respectively. The aim of our study was to bring novel biomechanical insights into these joint surgery procedures by investigating the influence of DoF at the TMC joint on muscle and joint forces in the thumb. A musculoskeletal model of the thumb was developed to equilibrate a 1 N external force in various directions while the thumb assumed key and pulp pinch postures. Muscle and joint forces were computed with an optimization method. In comparison to that of the 2-DoF (intact joint) condition, muscle forces slightly decreased in the 0-DoF (arthrodesis) condition, but drastically increased in the 3-DoF (arthroplasty) condition. TMC joint forces in the 3-DoF condition were 12 times larger than the 2-DoF joint. This study contributes to a further understanding of the biomechanics of the intact and surgically repaired TMC joint and addresses the biomechanical consequences of changing a joint's DoF by surgery.

Determination of passive moment-angle relationships at the trapeziometacarpal joint.

While modeling the trapeziometacarpal (TMC) joint for determination of tendon forces, the TMC has been considered frictionless and passive moments created by soft tissues neglected. This, however, becomes inaccurate when reaching the joint end range of motion and considering that the TMC is entirely crossed by a complex network of skin, ligaments, soft tissues, and tendons. The objective of this study was to evaluate the passive moments with respect to joint posture in order to further include this relationship in biomechanical modeling. An experimental method was proposed to estimate in vivo a global passive moment including the sum of the actions of each passive anatomical structure. An external force was applied at the level of the metacarpophalangeal joint in various directions ranging from neutral position to full extension and full adduction to full abduction. The passive moment was computed and expressed as a function of the adopted joint angles. An exponential regression was then developed to fit the experimental data and to propose a generic passive moment model. Results showed a good agreement between the proposed exponential regression model and the experimental measures. Moreover, it was shown that joint stiffness could represent more than 60% of the net joint moment during a typical pulp grip task. These results showed the necessity to include the data in biomechanical modeling. The results may help predict more realistic tendons force especially in abduction/adduction muscles.

Comparison of tendon tensions estimated from two biomechanical models of the thumb.

Despite the paramount function of the thumb in daily life, thumb biomechanical models have been little developed and studied. Moreover, only two studies provided quantitative anthropometric data of tendon moment arms. To investigate thumb tendon tensions, biomechanicians and clinicians have to know the performances and the limits of these two data sets. The aim of this study was thus to compare the results of these two models and evaluate their performances in regard to prior electromyographic measurements (EMG). Thumb posture was recorded during the classical key pinch and pulp pinch grips. Various fingertip forces applied at the distal segment were simulated in a range including extension, adduction, flexion, abduction. Input data of thumb postures and fingertip forces were used to compute tendon tensions with both models. Tendon tensions obtained using these two models were then compared and correlated to EMG measurements provided in the literature. The results showed that both models predicted relevant muscle coordination for five of the nine muscles modelled. Opponent and abductor longus muscle coordinations were badly estimated by both models. Each model was sensible to kinematic errors but not in the same proportion. This study pointed out the advantages/limits of the two models to use them more appropriately for clinical and/or research purposes.