A diagonal landing task to assess dynamic postural stability in ACL reconstructed females

A new method applied for explaining the landing patterns: Interpretability analysis of machine learning

As one of many fundamental sports techniques, the landing maneuver is also frequently used in clinical injury screening and diagnosis. However, the landing patterns are different under different constraints, which will cause great difficulties for clinical experts in clinical diagnosis. Machine learning (ML) have been very successful in solving a variety of clinical diagnosis tasks, but they all have the disadvantage of being black boxes and rarely provide and explain useful information about the reasons for making a particular decision. The current work validates the feasibility of applying an explainable ML (XML) model constructed by Layer-wise Relevance Propagation (LRP) for landing pattern recognition in clinical biomechanics. This study collected 560 groups landing data. By incorporating these landing data into the XML model as input signals, the prediction results were interpreted based on the relevance score (RS) derived from LRP. The interpretation obtained from XML was evaluated comprehensively from the statistical perspective based on Statistical Parametric Mapping (SPM) and Effect Size. The RS has excellent statistical characteristics in the interpretation of landing patterns between classes, and also conforms to the clinical characteristics of landing pattern recognition. The current work highlights the applicability of XML methods that can not only satisfy the traditional decision problem between classes, but also largely solve the lack of transparency in landing pattern recognition. We provide a feasible framework for realizing interpretability of ML decision results in landing analysis, providing a methodological reference and solid foundation for future clinical diagnosis and biomechanical analysis.

The influence of jump-landing direction on dynamic postural stability following anterior cruciate ligament reconstruction

BACKGROUND

Traditional testing prior to return to sport following anterior cruciate ligament reconstruction typically involves jump-landing tasks in the forward direction. As injury is most likely the result of multiplanar neuromuscular control deficits, assessment of dynamic postural stability using landing tasks that require multiplanar stabilization may be more appropriate. The purpose of this study was to examine how dynamic postural stability is affected when performing jump-landing tasks in three different directions.

METHODS

Fifteen athletes [11 females (18.0 ± 3.0 years) and 4 males (18.5 ± 3.1 years)] following anterior cruciate ligament reconstruction performed a series of single-limb jump-landing tasks in 3 directions. Individual directional stability indices and a composite dynamic postural stability index were calculated using ground reaction force data and were compared using separate one-way repeated measures ANOVAs.

FINDINGS

All directional stability indices demonstrated a significant main effect for jump-landing direction (medial-lateral P < 0.001, η2p = 0.95; anterior-posterior P < 0.001, η2p = 0.97; vertical P = 0.021, η2p = 0.24). The diagonal jump-landing direction produced increased medial-lateral stability and vertical stability scores, while the forward and diagonal jump-landing directions produced increased anterior-posterior stability scores. There was no significant effect for the composite dynamic stability index score.

INTERPRETATION

Jump-landing direction affects dynamic postural stability in all 3 planes of movement in athletes following anterior cruciate ligament reconstruction. Results indicate the potential need to incorporate multiple jump-landing directions to better assess dynamic postural stability prior to return to sport.

Bilateral Deficits in Dynamic Postural Stability in Females Persist Years after Unilateral ACL Injury and Are Modulated by the Match between Injury Side and Leg Dominance

Previous research has documented brain plasticity after an anterior cruciate ligament (ACL) tear and suggests that these neural adaptations contribute to poorer motor control. Since both brain hemispheres show adaptations, we hypothesized that reduced dynamic stability occurs not only in the injured, but also the contralateral, uninjured leg. Further, given brain hemispheric specialization's impact on motor coordination, we hypothesized the need to consider the injury side. A total of 41 female athletes and 18 controls performed single-leg jump-landings. Dynamic postural stability was measured as time-to-stabilization (TTS). We found reduced medio-lateral dynamic stability for the ACL injured leg (p = 0.006) with a similar trend for the contralateral leg (p = 0.050) compared to the control group. However, when distinguishing between injuries to the dominant and non-dominant legs, we found increased medio-lateral TTS only if the injury had occurred on the dominant side where landings on injured (p = 0.006) and contralateral (p = 0.036) legs required increased TTS. Assessments of dynamic stability, e.g., in the context of return-to-sport, should consider the injury side and compare results not only between the injured and the contralateral leg, but also to uninjured controls. Future research should not pool data from the dominant-leg ACL with non-dominant-leg ACL injuries when assessing post-injury motor performance.

Relationship of strength, joint kinesthesia, and plantar tactile sensation to dynamic and static postural stability among patients with anterior cruciate ligament reconstruction

Objective: Postural stability is essential for high-level physical activities after anterior cruciate ligament reconstruction (ACLR). This study was conducted to investigate the relationship of muscle strength, joint kinesthesia, and plantar tactile sensation to dynamic and static postural stability among patients with anterior cruciate ligament reconstruction. Methods: Forty-four patients over 6 months post anterior cruciate ligament reconstruction (age: 27.9 ± 6.8 years, height: 181.7 ± 8.7 cm, weight: 80.6 ± 9.4 kg, postoperative duration: 10.3 ± 3.6 months) participated in this study. Their static and dynamic postural stability, muscle strength, hamstring/quadriceps ratio, joint kinesthesia, and plantar tactile sensation were measured. Partial correlations were used to determine the correlation of the above-mentioned variables with time to stabilization (TTS) and root mean square of the center of pressure (COP-RMS) in anterior-posterior (AP) and mediolateral (ML) directions. Results: Both TTS_AP and TTS_ML were related to muscle strength and joint kinesthesia of knee flexion and extension; COP-RMS_AP was correlated with plantar tactile sensations at great toe and arch, while COP-RMS_ML was correlated with joint kinesthesia of knee flexion, and plantar tactile sensation at great toe and heel. Dynamic stability was sequentially correlated with strength and joint kinesthesia, while static stability was sequentially correlated with plantar tactile sensation and joint kinesthesia. Conclusion: Among patients with anterior cruciate ligament reconstruction, strength is related to dynamic postural stability, joint kinesthesia is related to dynamic and static postural stability, and plantar tactile sensation is related to static postural stability. Strength has a higher level of relationship to dynamic stability than joint kinesthesia, and plantar tactile sensation has a higher level of relationship to static stability than joint kinesthesia.

Kinetic measurement system use in individuals following anterior cruciate ligament reconstruction: a scoping review of methodological approaches

Purpose

Our primary objectives were to (1) describe current approaches for kinetic measurements in individuals following anterior cruciate ligament reconstruction (ACLR) and (2) suggest considerations for methodological reporting. Secondarily, we explored the relationship between kinetic measurement system findings and patient-reported outcome measures (PROMs).

Methods

We followed the PRISMA extension for scoping reviews and Arksey and O’Malley’s 6-stage framework. Seven electronic databases were systematically searched from inception to June 2020. Original research papers reporting parameters measured by kinetic measurement systems in individuals at least 6-months post primary ACLR were included.

Results

In 158 included studies, 7 kinetic measurement systems (force plates, balance platforms, pressure mats, force-measuring treadmills, Wii balance boards, contact mats connected to jump systems, and single-sensor insoles) were identified 4 main movement categories (landing/jumping, standing balance, gait, and other functional tasks). Substantial heterogeneity was noted in the methods used and outcomes assessed; this review highlighted common methodological reporting gaps for essential items related to movement tasks, kinetic system features, justification and operationalization of selected outcome parameters, participant preparation, and testing protocol details. Accordingly, we suggest considerations for methodological reporting in future research. Only 6 studies included PROMs with inconsistency in the reported parameters and/or PROMs.

Conclusion

Clear and accurate reporting is vital to facilitate cross-study comparisons and improve the clinical application of kinetic measurement systems after ACLR. Based on the current evidence, we suggest methodological considerations to guide reporting in future research. Future studies are needed to examine potential correlations between kinetic parameters and PROMs.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40634-021-00397-0.

Side-hops challenge knee control in the frontal and transversal plane more than hops for distance or height among ACL-reconstructed individuals

We compared knee landing mechanics with presumed relation to risk of anterior cruciate ligament (ACL) injury among three single-leg hop tests and between legs in individuals with unilateral ACL reconstruction. Thirty-four participants (>10 months' post-surgery, 23 females) performed the standardised rebound side hop (SRSH), maximal hop for distance (OLHD) and maximal vertical hop (OLVH). We calculated the following knee outcomes from motion capture and force plate data: finite helical axis inclination angles (approximates knee robustness), frontal and transversal plane angles at initial contact, peak angles of abduction and internal rotation during landing, and peak external moments of flexion, abduction and internal rotation during landing. Repeated-measures MANOVA analysis ('sex' as covariate) confirmed that SRSH induced greater angles and moments, particularly in the frontal plane, compared to OLHD and OLVH. There was between-leg asymmetry for peak knee flexion moment for males during OLHD and OLVH, and for females during SRSH. Our results advocate the SRSH over OLHD and OLVH for assessment of knee landing control to screen for movement patterns potentially related to ACL injury risk. However, clear differences in both knee kinematics and kinetics between OLHD and SRSH motivate the use of both tests to evaluate different aspects of landing control.

Postural control is altered in females with excessive medial knee displacement

Knee valgus motion observed during landing tasks has been proposed as a predictor of future knee injury. It mainly involves excess motion in the frontal plane and is known to be greater in individuals with excessive medial knee displacement (MKD). This affects postural control during sports manoeuvres. Previous sports medicine-related research suggests that the nature of these fluctuations provide rich and more sensitive information to identify risk of (re)injury. We aimed to investigate the fluctuations of the centre of pressure (CoP) in individuals with and without excessive MKD. Twenty females (12 controls; 8 excessive MKD) were instructed to perform single-leg landing tasks from three different directions. The participants landed on a force plate and stayed still for 20 seconds. The fluctuations of the anterior-posterior and medial-lateral directions of the CoP were determined through the calculation of Sample Entropy. Mixed-model ANOVAs (3 [Landing Direction] x 2 [Group]) were used. We have found that only the entropy of the medial-lateral direction was different between groups. Individuals with excessive MKD exhibited an increase in entropy values, indicating greater randomness in CoP fluctuations. This suggests a decreased ability to adapt to environmental demands that likely result in an increased risk of injury.

Anterior cruciate ligament reconstruction and dynamic stability at time of release for return to sport

OBJECTIVE

Examine dynamic stability using Dynamic Postural Stability Index (DPSI) in athletes following anterior cruciate ligament reconstruction (ACLR) at time of release for return-to-sport (RTS), compared to matched controls.

DESIGN

Cross-sectional case-control study.

SETTING

Sports medicine clinic.

SUBJECTS

Fifteen ACLR athletes who had completed post-operative rehabilitation and were within 6 weeks following release to RTS were age-, gender-, and activity-matched to 15 healthy controls.

MAIN OUTCOME MEASURES

Ground reaction forces (GRFs) were collected using a portable force plate during stabilization from three different single-leg landing tasks. A composite DPSI was calculated using GRFs.

RESULTS

Compared to matched controls, ACLR athletes within 6 weeks of release for RTS did not significantly differ in dynamic postural stability and there were no significant differences between the involved and uninvolved limbs in the ACLR group.

CONCLUSION

Current findings indicate that dynamic postural stability, as measured using the DPSI, is not significantly different in ACLR subjects at time of release for RTS compared to matched controls. In addition, the DPSI was not significantly different between the involved and uninvolved limbs in the ACLR subjects. The results suggest that the post-ACLR rehabilitation program utilized may have adequately restored postural stability in this particular sample.

Sprint mechanics return to competition follow-up after hamstring injury on a professional soccer player: A case study with an inertial sensor unit based methodological approach

BACKGROUND

The present research aimed to describe an inertial unit (IU)-based sprint mechanics evaluation model for assessing players' readiness to return to competition after suffering a grade I hamstring injury.

METHODS

A professional male football player (age 19years; height 177cm; weight 70kg, midfielder, Spanish, 3° Division) with a grade 1 biceps femoris injury was evaluated at pre-season, at return to play after injury and at the end of the competitive season. Sprint mechanics were analyzed via the use of an inertial orientation tracker (Xsens Technologies B.V. Enschede, Netherlands) attached over the L3-L4 region of the subject's lumbar spine. Sprint mechanics such as horizontal components of ground reaction force were assessed in both legs during sprinting actions. Findings and interpretation: Both the coefficient of the horizontal force application (SFV) and the ratio of forces (DRF) applied at increasing velocity were decreased in the injured limb compared with the contralateral healthy limb at the return to play evaluation (73% and 76% reductions, respectively) and returned to symmetrical levels at the end-season evaluation.

An Ecological Study of Anterior Cruciate Ligament Reconstruction, Part 2: Functional Performance Tests Correlate With Return-to-Sport Outcomes

Background:

Additional high-quality prospective studies are needed to better define the objective criteria used in relation to return-to-sport decisions after anterior cruciate ligament (ACL) reconstruction in active populations.

Purpose:

To investigate prospectively the relationship between functional performance test results at 24 weeks postoperative and return-to-sport activity (Tegner activity score) at 12 and 24 months, respectively, after synthetic (ligament advanced reinforcement system [LARS]) and autograft (doubled semitendinosus/gracilis [2ST/2GR]) ACL reconstructions.

Study Design:

Case series; Level of evidence, 4.

Methods:

A total of 64 patients who underwent ACL reconstruction (32 LARS, 32 2ST/2GR autograft; mean age, 27.9 years; body mass index [BMI], 24.9 kg/m²) were assessed preoperatively and at staged intervals postoperatively up to 24 weeks for isokinetic testing of quadriceps and hamstring average power per repetition at 60 deg/s and 180 deg/s, a battery of hop tests, peak vertical ground-reaction force (vGRF), and time to peak vGRF (in seconds) during a step- and jump-down task onto a force platform and peak speed (m/s) using a global positioning system (GPS unit) during a running task. A cohort of 32 healthy matched participants (mean age, 26.31 years; BMI, 25.7 kg/m²) were also tested to act as reference. Pearson correlation was calculated to assess correlation of each performance measure at 24 weeks postoperative with activity outcomes (Tegner score) at 12 and 24 months.

Results:

The strongest correlation between physical performance tests and return-to-sport outcomes was observed with peak speed during running. Large correlations were also observed for hamstring isokinetic power and hop test for distance. Moderate correlations were observed for timed hop, peak vGRF during a jump-down task, and quadriceps isokinetic power. No statistical correlations were observed for time to peak vGRF during a step-down and jump-down task as well as peak vGRF during a step-down task. When the performance tests were pooled together, mean postoperative improvements of 24% were observed from preoperative to 24 weeks within the surgical cohort. For each performance test, preoperative level of function strongly correlated with performance levels on the same test at 24 weeks.

Discussion:

The results of this study indicate that clinicians might seek to prioritize these tests and the rehabilitation themes they imply when seeking to maximize postoperative ACL activity outcomes. The observed strength between pre- and postoperative performance tests and return-to-sport outcomes within this study highlights the potential value of preoperative conditioning before undergoing ACL reconstruction. Future research should examine absolute predictive criterion thresholds for functional performance-based tests and reinjury risk reduction after ACL reconstruction.

Baseline Time to Stabilization Identifies Anterior Cruciate Ligament Rupture Risk in Collegiate Athletes

BACKGROUND

There is a need for successful screening methods to identify athletes at increased risk of anterior cruciate ligament (ACL) injury. Previous research showed that collegiate athletes with ACL tears demonstrated slower time to stabilization during jump landing after reconstruction.

HYPOTHESIS

Collegiate athletes with baseline deficiencies in time to stabilization are at increased risk of subsequent ACL rupture.

STUDY DESIGN

Case-control study; Level of evidence, 3.

METHODS

A total of 278 National Collegiate Athletic Association Division I college athletes (166 men, 112 women; mean age, 18.5 years; height, 178.8 cm; mass, 79.9 kg) in the high-risk sports of men's football; women's volleyball and field hockey; and men's and women's lacrosse, basketball, and soccer were measured to obtain baseline time to stabilization for backward, forward, medial, and lateral single-legged jump landing tasks. Athletes were followed for ACL rupture over a 4-year period. Independent t tests were used to evaluate differences in time to stabilization for each jump landing task between athletes with subsequent ACL rupture and uninjured athletes. Logistic regression models were used to assess time to stabilization as a predictor for ACL rupture.

RESULTS

Nine athletes sustained noncontact ACL ruptures (5 men, 4 women). These 9 athletes took significantly longer to stabilize compared with uninjured athletes during baseline backward jump landing (1.58 ± 0.39 and 1.09 ± 0.52 seconds, respectively; P = .0052). The odds of ACL rupture increased 3-fold (odds ratio, 2.95; 95% CI, 1.28-6.77) for every second increase in backward time to stabilization observed between injured and uninjured athletes.

CONCLUSION

Collegiate athletes with slower baseline backward time to stabilization were at increased risk of ACL rupture.

Vertical jumping biomechanical evaluation through the use of an inertial sensor-based technology

Progress in micro-electromechanical systems has turned inertial sensor units (IUs) into a suitable tool for vertical jumping evaluation. In total, 9 men and 8 women were recruited for this study. Three types of vertical jumping tests were evaluated in order to determine if the data provided by an IU placed at the lumbar spine could reliably assess jumping biomechanics and to examine the validity of the IU compared with force plate platform recordings. Robust correlation levels of the IU-based jumping biomechanical evaluation with respect to the force plate across the entire analysed jumping battery were found. In this sense, significant and extremely large correlations were found when raw data of both IU and force plate-derived normalised force-time curves were compared. Furthermore, significant and mainly moderate correlation levels were also found between both instruments when isolated resultant forces' peak values of predefined jumping phases of each manoeuvre were analysed. However, Bland and Altman graphical representation demonstrated a systematic error in the distribution of the data points within the mean ±1.96 SD intervals. Using IUs, several biomechanical variables such as the resultant force-time curve patterns of the three different vertical jumps analysed were reliably measured.

Postural stability deficits during the transition from double-leg stance to single-leg stance in anterior cruciate ligament reconstructed subjects

The goal of this study was to evaluate postural stability during the transition from double-leg stance (DLS) to single-leg stance (SLS) in anterior cruciate ligament reconstructed (ACLR) (n=20) and non-injured control subjects (n=20). All ACLR subjects had fully returned to their pre-injury sport participation. Both groups were similar for age, gender, height, weight, body mass index and activity level. Spatiotemporal center of pressure outcomes of both legs of each subject were measured during the transition from DLS to SLS in eyes open and eyes closed conditions. Movement speed was standardized. The center of pressure displacement after a new stability point was reached during the SLS phase was significantly increased in the ACLR group compared to the control group in the eyes closed condition (P=.001). No significant different postural stability outcomes were found between the operated and non-operated legs. In conclusion, the ACLR group showed postural stability deficits, indicating that these persons may have a decreased ability to stabilize their body after the internal postural perturbation created by the transition from DLS to SLS. The non-operated leg may not be the best reference when evaluating postural stability of the operated leg after ACLR, as no differences were found between legs.