Knee joint forces (KJF) are biomechanical measures used to infer the load on knee joint structures. The purpose of this study is to develop an artificial neural network (ANN) that estimates KJF during sport movements, based on data obtained by wearable sensors. Thirteen participants were equipped with two inertial measurement units (IMUs) located on the right leg. Participants performed a variety of movements, including linear motions, changes of direction, and jumps. Biomechanical modelling was carried out to determine KJF. An ANN was trained to model the association between the IMU signals and the KJF time series. The ANN-predicted KJF yielded correlation coefficients that ranged from 0.60 to 0.94 (vertical KJF), 0.64 to 0.90 (anterior–posterior KJF) and 0.25 to 0.60 (medial–lateral KJF). The vertical KJF for moderate running showed the highest correlation (0.94 ± 0.33). The summed vertical KJF and peak vertical KJF differed between calculated and predicted KJF across all movements by an average of 5.7% ± 5.9% and 17.0% ± 13.6%, respectively. The vertical and anterior–posterior KJF values showed good agreement between ANN-predicted outcomes and reference KJF across most movements. This study supports the use of wearable sensors in combination with ANN for estimating joint reactions in sports applications.

Joint moment measurements represent an objective biomechanical parameter of knee joint load in knee osteoarthritis (KOA). Wearable sensors in combination with machine learning techniques may provide solutions to develop assistive devices in KOA patients to improve disease treatment and to minimize risk of non-functional overreaching (e.g., pain). The purpose of this study was to develop an artificial neural network (ANN) that estimates external knee flexion moments (KFM) and external knee adduction moments (KAM) during various locomotion tasks, based on data obtained by two wearable sensors. Thirteen participants were instrumented with two inertial measurement units (IMUs) located on the right thigh and shank. Participants performed six different locomotion tasks consisting of linear motions and motions with a change of direction, while IMU signals as well as full body kinematics and ground reaction forces were synchronously recorded. KFM and KAM were determined using a full body biomechanical model. An ANN was trained to estimate the KFM and KAM time series using the IMU signals as input. Evaluation of the ANN was done using a leave-one-subject-out cross-validation. Concordance of the ANN-estimated KFM and reference data was categorized for five tasks (walking straight, 90° walking turn, moderate running, 90° running turn and 45° cutting maneuver) as strong (r ≥ 0.69, rRMSE ≤ 23.1) and as moderate for fast running (r = 0.65 ± 0.43, rRMSE = 25.5 ± 7.0%). For all locomotion tasks, KAM yielded a lower concordance in comparison to the KFM, ranging from weak (r ≤ 0.21, rRMSE ≥ 33.8%) in cutting and fast running to strong (r = 0.71 ± 0.26, rRMSE = 22.3 ± 8.3%) for walking straight. Smallest mean difference of classical discrete load metrics was seen for KFM impulse, 10.6 ± 47.0%. The results demonstrate the feasibility of using only two IMUs to estimate KFM and KAM to a limited extent. This methodological step facilitates further work that should aim to improve the estimation accuracy to provide valuable biofeedback systems for KOA patients. Greater accuracy of effective implementation could be achieved by a participant- or task-specific ANN modeling.

Skating is a fundamental movement in ice hockey; however little research has been conducted within the field of hockey skating biomechanics due to the difficulties of on-ice data collection. In this study a novel on-ice measurement approach was tested for reliability, and subsequently implemented to investigate the forward skating technique, as well as technique differences across skill levels. Nine high caliber (High) and nine low caliber (Low) hockey players performed 30m forward skating trials. A 3D accelerometer was mounted to the right skate for the purpose of stride detection, with the 2nd and 6th strides defined as acceleration and steady-state, respectively. The activity of five lower extremity muscles was recorded using surface electromyography. Biaxial electro-goniometers were used to quantify hip and knee angles, and in-skate plantar force was measured using instrumented insoles. Reliability was assessed with the coefficient of multiple correlation, which demonstrated moderate (r>0.65) to excellent (r>0.95) scores across selected measured variables. Greater plantar-flexor muscle activity and hip extension were evident during acceleration strides, while steady state strides exhibited greater knee extensor activity and hip abduction range of motion (p<0.05). High caliber exhibited greater hip range of motion and forefoot force application (p<0.05). The successful implementation of this on-ice mobile measurement approach offers potential for athlete monitoring, biofeedback and training advice.

The capturing of movements by means of wearable sensors has become increasingly popular in order to obtain sport performance measures during training or competition. The purpose of the current study was to investigate the feasibility of using body worn accelerometers to identify previous highlighted performance related biomechanical changes in terms of substantial differences across skill levels and skating phases. Twenty‐two ice hockey players of different caliber were equipped with two 3D accelerometers, located on the skate and the waist, as they performed 30 m forward skating sprints on an ice rink. Two measures of the temporal stride characteristics (contact time and stride time) and one measure of the propulsive power (stride propulsion) of a skating stride were calculated and checked for discriminating effects across (i) skill levels and (ii) sprint phases as well as for their (iii) strength of association with the sprint performance (total sprint time). High caliber players showed an increased stride propulsion (+22%, P < 0.05) and shorter contact time (−5%, P < 0.05). All three analysed variables highlighted substantial biomechanical differences between the accelerative and constant velocity phases (P < 0.05). Stride propulsion of acceleration strides primarily correlated to total sprint time (r = −0.57, P < 0.05). The results demonstrate the potential of accelerometers to assess skating technique elements such as contact time or elements characterizing the propulsive power such as center of mass acceleration, to gauge skating performance. Thus, the findings of this study might contribute to establishing wearable sensors for in‐field ice hockey skating performance analysis.

Despite good clinical functional outcome, deficits in gait biomechanics exist 2 years after total hip replacement surgery. The aims of this research were (1) to group patients showing similar gait adaptations to hip osteoarthritis and (2) to investigate the effect of the surgical treatment on gait kinematics and external joint moments. In a secondary analysis, gait data of 51 patients with unilateral hip osteoarthritis were analyzed. A k-means cluster analysis was performed on scores derived via a principal component analysis of the gait kinematics. Preoperative and postoperative datasets were statistically tested between clusters and 46 healthy controls. The first three principal components incorporated hip flexion/extension, pelvic tilt, foot progression angle and thorax tilt. Two clusters were discriminated best by the peak hip extension during terminal stance. Both clusters deviated from healthy controls in spatio-temporal, kinematic and kinetic parameters. The cluster with less hip extension deviated significantly more. The clusters improved postoperatively but differences to healthy controls were still present one year after surgery. A poor preoperative gait pattern in patients with unilateral hip osteoarthritis is associated with worse gait kinematics after total hip replacement. Further research should focus on the identification of patients who can benefit from an adapted or individualized rehabilitation program.