Matthew Thomas Owen Worsey
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Matthew Thomas Owen Worsey has a total of 71 co-authors. Here are the top co-authors by number of shared publications:
- David Victor Thiel
School of Engineering and Built Environment, Griffith University, Brisbane, QLD, Australia
15 shared publications - Hugo G. Espinosa
School of Engineering and Built Environment, Griffith University, Brisbane, Australia
15 shared publications - Jonathan Bruce Shepherd
Griffith School of Engineering and Built Environment, Griffith University, Nathan Campus, Queensland 4111, Australia
8 shared publications - David G. Lloyd
Australian Centre for Precision Health and Technology (PRECISE), Griffith University
3 shared publications Claudio Pizzolato
2 shared publications- Claudio Pizzolato
Griffith University
2 shared publications
Latest Publications
- Open Access
- 28 July 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Leveraging inertial sensors and dynamic time warping to identify biomarkers for gait impairment in hereditary spastic paraplegia: an exploratory study
Hereditary Spastic Paraplegia (HSP) is a rare neurodegenerative disorder characterized by progressive spasticity and weakness of the lower limbs, associated with gait impairment. While inertial measurement units (IMUs) have shown promise in monitoring mobility, objective biomechanical biomarkers of gait dysfunction remain underdeveloped. This study aimed to evaluate whether normalized root mean square error (nRMSE), calculated after Dynamic Time Warping (DTW) alignment of gyroscopic gait cycle profiles, could serve as a candidate biomarker for gait impairment in HSP. We conducted a cross-sectional pilot study involving six individuals with clinically diagnosed HSP and six broadly comparable healthy controls. Participants walked along a nine-meter walkway under standardized conditions, wearing two IMUs on the shanks. Sagittal plane angular velocity was extracted from each gait cycle. Gait parameters were computed, and each participant’s mean gait cycle was aligned to the control mean using DTW. Deviations were quantified using nRMSE. Clinical assessments included the SPATAX-EUROSPA Disability Score and observable items from the Spastic Paraplegia Rating Scale (SPRS). HSP participants showed longer stance times, slower speeds, and higher nRMSE scores. nRMSE was strongly correlated with SPATAX ( R = 0.967) and SPRS ( R = 0.982) in linear regression models. These findings support the potential utility of nRMSE as an objective, sensor-derived marker of gait impairment in HSP, warranting further validation in larger and longitudinal cohorts.
- 28 July 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Open AccessLeveraging inertial sensors and dynamic time warping to identify biomarkers for gait impairment in hereditary spastic paraplegia: an exploratory study
Hereditary Spastic Paraplegia (HSP) is a rare neurodegenerative disorder characterized by progressive spasticity and weakness of the lower limbs, associated with gait impairment. While inertial measurement units (IMUs) have shown promise in monitoring mobility, objective biomechanical biomarkers of gait dysfunction remain underdeveloped. This study aimed to evaluate whether normalized root mean square error (nRMSE), calculated after Dynamic Time Warping (DTW) alignment of gyroscopic gait cycle profiles, could serve as a candidate biomarker for gait impairment in HSP. We conducted a cross-sectional pilot study involving six individuals with clinically diagnosed HSP and six broadly comparable healthy controls. Participants walked along a nine-meter walkway under standardized conditions, wearing two IMUs on the shanks. Sagittal plane angular velocity was extracted from each gait cycle. Gait parameters were computed, and each participant’s mean gait cycle was aligned to the control mean using DTW. Deviations were quantified using nRMSE. Clinical assessments included the SPATAX-EUROSPA Disability Score and observable items from the Spastic Paraplegia Rating Scale (SPRS). HSP participants showed longer stance times, slower speeds, and higher nRMSE scores. nRMSE was strongly correlated with SPATAX ( R = 0.967) and SPRS ( R = 0.982) in linear regression models. These findings support the potential utility of nRMSE as an objective, sensor-derived marker of gait impairment in HSP, warranting further validation in larger and longitudinal cohorts.
- Open Access
Real-time hip biomechanics from smart garments via a physics-informed neural network
Tissue-level mechanical stimuli are primary drivers of tissue adaptation and can be optimised during conservative treatments to improve treatment outcomes for many highly prevalent musculoskeletal conditions. Current laboratory-based technologies limit our ability to connect conservative interventions such as exercise and movement modification with muscle, joint, and tissue-level mechanics, in natural environments. We introduce a physics-informed neural network (PINN) to estimate clinically relevant biomechanics from smart garments. By accounting for physiological dynamics of neural activation and muscle contraction, the PINN accurately predicted hip joint angles (RMSE <6 degrees), moments (RMSE 0.12 N·m/kg to 0.30 N·m/kg), and joint forces (RMSE 6 to 16%) from three inertial measurement units and four electromyographic sensors. We demonstrated that the trained PINN can be combined with a smart garment to estimate hip biomechanics, in real-time, during a gait retraining intervention aimed at modifying joint loading to treat hip osteoarthritis. The developed PINN and smart garment system may be adapted and generalised for personalised management or rehabilitation of a broad range of musculoskeletal diseases and injuries, in clinical, home, workplace, and sporting environments.
Open AccessReal-time hip biomechanics from smart garments via a physics-informed neural network
Tissue-level mechanical stimuli are primary drivers of tissue adaptation and can be optimised during conservative treatments to improve treatment outcomes for many highly prevalent musculoskeletal conditions. Current laboratory-based technologies limit our ability to connect conservative interventions such as exercise and movement modification with muscle, joint, and tissue-level mechanics, in natural environments. We introduce a physics-informed neural network (PINN) to estimate clinically relevant biomechanics from smart garments. By accounting for physiological dynamics of neural activation and muscle contraction, the PINN accurately predicted hip joint angles (RMSE <6 degrees), moments (RMSE 0.12 N·m/kg to 0.30 N·m/kg), and joint forces (RMSE 6 to 16%) from three inertial measurement units and four electromyographic sensors. We demonstrated that the trained PINN can be combined with a smart garment to estimate hip biomechanics, in real-time, during a gait retraining intervention aimed at modifying joint loading to treat hip osteoarthritis. The developed PINN and smart garment system may be adapted and generalised for personalised management or rehabilitation of a broad range of musculoskeletal diseases and injuries, in clinical, home, workplace, and sporting environments.
- Open Access
- 10 August 2025
- Published by Elsevier in Journal of Biomechanics
- Vol. 191, 112890
Muscle forces and powers during sprint cycling at optimal cadence
- 10 August 2025
- Published by Elsevier in Journal of Biomechanics
- Vol. 191, 112890
Open AccessMuscle forces and powers during sprint cycling at optimal cadence
- 22 April 2025
- Published by SAGE Publications in International Journal of Sports Science & Coaching
The interplay between anthropometric profile, surfboard dimensions, and surfboard design and selection processes in competitive female and male surfers
Surfboards are a critical component of surfing performance, yet the relationship between a surfer's anthropometric profile and their surfboard dimensions, as well as the factors influencing surfers’ decision-making around surfboard design and selection within competition, remains unexplored. Forty-two competitive surfers participated in this study. Of the forty-two surfers, the anthropometric profile of twenty-eight females (n = 12) and males (n = 16) was determined (i.e., height, body mass, arm span, body composition), while thirty-one females (n = 15) and males (n = 16) completed a survey to determine their decision-making towards surfboard design and selection within competition. The results demonstrated fewer significant correlations between the anthropometric profiles and surfboard dimensions of female surfers than male surfers and the anthropometric profiles of female surfers explained less variance in their surfboard dimensions compared to male surfers. Fat mass measures contributed more to the variance in female surfers’ surfboard dimensions, whereas lean mass measures contributed more to the variance for male surfers. Survey results also determined that female surfers rated the factors informing their decision-making around surfboard design and selection as less influential compared to male surfers. These findings highlight notable sex-related differences in the relationship between surfers’ anthropometric profile and their competition surfboards, as well as their decision-making processes. Increased exposure and experimentation with surfboard design could benefit female surfers and advance their performance in the sport alongside their male counterparts.
- 22 April 2025
- Published by SAGE Publications in International Journal of Sports Science & Coaching
The interplay between anthropometric profile, surfboard dimensions, and surfboard design and selection processes in competitive female and male surfers
Surfboards are a critical component of surfing performance, yet the relationship between a surfer's anthropometric profile and their surfboard dimensions, as well as the factors influencing surfers’ decision-making around surfboard design and selection within competition, remains unexplored. Forty-two competitive surfers participated in this study. Of the forty-two surfers, the anthropometric profile of twenty-eight females (n = 12) and males (n = 16) was determined (i.e., height, body mass, arm span, body composition), while thirty-one females (n = 15) and males (n = 16) completed a survey to determine their decision-making towards surfboard design and selection within competition. The results demonstrated fewer significant correlations between the anthropometric profiles and surfboard dimensions of female surfers than male surfers and the anthropometric profiles of female surfers explained less variance in their surfboard dimensions compared to male surfers. Fat mass measures contributed more to the variance in female surfers’ surfboard dimensions, whereas lean mass measures contributed more to the variance for male surfers. Survey results also determined that female surfers rated the factors informing their decision-making around surfboard design and selection as less influential compared to male surfers. These findings highlight notable sex-related differences in the relationship between surfers’ anthropometric profile and their competition surfboards, as well as their decision-making processes. Increased exposure and experimentation with surfboard design could benefit female surfers and advance their performance in the sport alongside their male counterparts.
- Open Access
- 18 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Improving Calibration of EMG-Informed Neuromusculoskeletal Models Through Differentiable Physics and Muscle Synergies
Electromyogram (EMG)-informed neuromusculoskeletal (NMS) models can predict physiologically plausible muscle forces and joint moments. However, calibrating model parameters (e.g., optimal fiber length, tendon slack length) to the individual is time-consuming, with the optimization often requiring hours to converge and typically not accounting for unrecorded muscle excitations. This study addresses these limitations by incorporating differentiable physics and muscle synergies into the calibration of NMS models. Methods: We implemented an NMS model with auto-differentiable Hill-type muscles, enabling the use of adaptive gradient descent optimizers. Two types of calibration were evaluated: a standard EMG-driven approach and a synergy-hybrid approach that also synthesized unrecorded excitations. These methods were evaluated using upper and lower limb data, each from a single participant. Results: The calibration time was reduced by up to 26 times while maintaining comparable accuracy in moment predictions. Compared to the EMG-driven calibration, the synergy-hybrid calibration improved the estimates of model parameters for reduced number of EMG channels. Conclusion: Autodifferentiable Hill-type muscle models greatly reduce NMS model calibration time and enables the synthesis of unrecorded muscle excitations through muscle synergies, facilitating the calibration of all muscle parameters. Significance: This new rapid calibration could support deployment of NMS models in time-sensitive applications, including real-time biomechanical analyses and personalized neurorehabilitation.
- 18 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Open AccessImproving Calibration of EMG-Informed Neuromusculoskeletal Models Through Differentiable Physics and Muscle Synergies
Electromyogram (EMG)-informed neuromusculoskeletal (NMS) models can predict physiologically plausible muscle forces and joint moments. However, calibrating model parameters (e.g., optimal fiber length, tendon slack length) to the individual is time-consuming, with the optimization often requiring hours to converge and typically not accounting for unrecorded muscle excitations. This study addresses these limitations by incorporating differentiable physics and muscle synergies into the calibration of NMS models. Methods: We implemented an NMS model with auto-differentiable Hill-type muscles, enabling the use of adaptive gradient descent optimizers. Two types of calibration were evaluated: a standard EMG-driven approach and a synergy-hybrid approach that also synthesized unrecorded excitations. These methods were evaluated using upper and lower limb data, each from a single participant. Results: The calibration time was reduced by up to 26 times while maintaining comparable accuracy in moment predictions. Compared to the EMG-driven calibration, the synergy-hybrid calibration improved the estimates of model parameters for reduced number of EMG channels. Conclusion: Autodifferentiable Hill-type muscle models greatly reduce NMS model calibration time and enables the synthesis of unrecorded muscle excitations through muscle synergies, facilitating the calibration of all muscle parameters. Significance: This new rapid calibration could support deployment of NMS models in time-sensitive applications, including real-time biomechanical analyses and personalized neurorehabilitation.
- Open Access
- 21 February 2025
- Published by Taylor & Francis in Journal of Sports Sciences
- Vol. 43 (5) , 512-522
Relationship between the cervical spine and impacts whilst heading the ball in adolescent male soccer players
There has been considerable public concern regarding the effects of repetitive head impacts associated with heading the ball in soccer. This is relevant for adolescents whose brains are undergoing rapid development. The study investigated the relationship between strength, proprioception, endurance and range of motion of the cervical spine and impacts associated with heading the ball in adolescent male soccer players. The participants were 20 skilled players aged 15-18 years. Outcome measures included questionnaires, assessments of cervical spine function (range of motion, joint position error, isometric muscle strength and endurance) and the ratio of maximum head acceleration to maximum T3 acceleration (impact ratio) during each ball impact when heading the ball. Results showed that increased range of motion of extension of the cervical spine and lower joint position error in flexion of the cervical spine were associated with higher impact ratios. Greater isometric strength of the lateral flexors was associated with lower impact ratios. Endurance of the deep neck flexors was not associated with impact ratios. Strength training may be beneficial to decrease head impacts during soccer ball heading, especially in those with increased range of motion of extension of the cervical spine.
- 21 February 2025
- Published by Taylor & Francis in Journal of Sports Sciences
- Vol. 43 (5) , 512-522
Open AccessRelationship between the cervical spine and impacts whilst heading the ball in adolescent male soccer players
There has been considerable public concern regarding the effects of repetitive head impacts associated with heading the ball in soccer. This is relevant for adolescents whose brains are undergoing rapid development. The study investigated the relationship between strength, proprioception, endurance and range of motion of the cervical spine and impacts associated with heading the ball in adolescent male soccer players. The participants were 20 skilled players aged 15-18 years. Outcome measures included questionnaires, assessments of cervical spine function (range of motion, joint position error, isometric muscle strength and endurance) and the ratio of maximum head acceleration to maximum T3 acceleration (impact ratio) during each ball impact when heading the ball. Results showed that increased range of motion of extension of the cervical spine and lower joint position error in flexion of the cervical spine were associated with higher impact ratios. Greater isometric strength of the lateral flexors was associated with lower impact ratios. Endurance of the deep neck flexors was not associated with impact ratios. Strength training may be beneficial to decrease head impacts during soccer ball heading, especially in those with increased range of motion of extension of the cervical spine.
- Open Access
- 16 April 2023
- Published by Elsevier in Journal of Science and Medicine in Sport
- Vol. 26, S30-S39
Maintaining soldier musculoskeletal health using personalised digital humans, wearables and/or computer vision
- 16 April 2023
- Published by Elsevier in Journal of Science and Medicine in Sport
- Vol. 26, S30-S39
Open AccessMaintaining soldier musculoskeletal health using personalised digital humans, wearables and/or computer vision
- 20 October 2021
- Published by Springer Nature in Sports Engineering
- Vol. 24 (1) , 1-15
Automatic classification of running surfaces using an ankle-worn inertial sensor
- 20 October 2021
- Published by Springer Nature in Sports Engineering
- Vol. 24 (1) , 1-15
Automatic classification of running surfaces using an ankle-worn inertial sensor
- 10 September 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 70 (00189456) , 1-11
Surface Electric Field Variations Due to Different Human Postures for Wireless Endoscopy Applications
RSSI based localization methods for wireless capsule endoscopy (WCE) application requires a patient to wear on-body sensors detecting the RF signals from the capsule. The main advantage of WCE is to allow the patient to live normally and to move freely during the natural transit time, which could take up to 73 hours. Changes in posture can be accompanied by the movement of soft tissues in the abdominal region. This will affect the propagation path length between on-body receiving sensors and the ingested wireless capsule, and consequently the received signal strength (RSS). This paper explores potential variations in RSSI measurements created by postural change. Retro-reflective markers placed in direct contact with the skin of a participant were used to represent appropriate locations of the on-body slot antennas, which could be used for WCE localization. Optical motion capture technology was used to determine an estimate of the location of the markers and corresponding skin position for different postures. The relative displacement of the markers from their initial location (the standing posture) was a maximum of 24 mm for the lying posture (supine position) for the participant with BMI of 29 kg/m2. The maximum variation of the surface electric field of 13.77 dB was calculated using the attenuation path loss (PL) model for the human abdominal region propagation environment at 2.45 GHz. If not anticipated, posture induced variations in RSSI can produce misleading interpretations, and will require adjustments of WCE localization algorithms.
- 10 September 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 70 (00189456) , 1-11
Surface Electric Field Variations Due to Different Human Postures for Wireless Endoscopy Applications
RSSI based localization methods for wireless capsule endoscopy (WCE) application requires a patient to wear on-body sensors detecting the RF signals from the capsule. The main advantage of WCE is to allow the patient to live normally and to move freely during the natural transit time, which could take up to 73 hours. Changes in posture can be accompanied by the movement of soft tissues in the abdominal region. This will affect the propagation path length between on-body receiving sensors and the ingested wireless capsule, and consequently the received signal strength (RSS). This paper explores potential variations in RSSI measurements created by postural change. Retro-reflective markers placed in direct contact with the skin of a participant were used to represent appropriate locations of the on-body slot antennas, which could be used for WCE localization. Optical motion capture technology was used to determine an estimate of the location of the markers and corresponding skin position for different postures. The relative displacement of the markers from their initial location (the standing posture) was a maximum of 24 mm for the lying posture (supine position) for the participant with BMI of 29 kg/m2. The maximum variation of the surface electric field of 13.77 dB was calculated using the attenuation path loss (PL) model for the human abdominal region propagation environment at 2.45 GHz. If not anticipated, posture induced variations in RSSI can produce misleading interpretations, and will require adjustments of WCE localization algorithms.
- Open Access
- 18 February 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Letters
- Vol. 5 (3) , 1-4
One Size Doesn't Fit All: Supervised Machine Learning Classification in Athlete-Monitoring
Athlete movement data is integral for optimizing athlete-performance and can lead to reduced fatigue and in turn can mitigate injury risk. There is a substantial amount of scientific literature which investigates the ability of computer-vision and inertial sensor technologies to classify sport-specific movements. The coupling of automatic sport action labelling and athlete-monitoring data can significantly enhance athlete work-load monitoring. Two recent systematic reviews of the literature, pertinent to sport-specific movement classification, revealed that the majority of journal articles use athlete-dependent classification model training and evaluation methods. These methods can significantly enhance model classification performance, particularly with movements which have high inter-athlete technique variation. This is because it enables models to learn features distinctive to all athletes during training. This manuscript details the training and evaluation of supervised machine learning models to automatically classify running surface (athletics track, hard sand and soft sand) using features extracted from an upper-back inertial measurement unit sensor. Possible classification performance enhancement is demonstrated by comparing athlete independent and athlete dependent supervised machine learning methods. Using athlete dependent methods significantly increased the classification performance in terms of weighted average precision, recall, F1-score and accuracy (p < 0.05).
- 18 February 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Letters
- Vol. 5 (3) , 1-4
Open AccessOne Size Doesn't Fit All: Supervised Machine Learning Classification in Athlete-Monitoring
Athlete movement data is integral for optimizing athlete-performance and can lead to reduced fatigue and in turn can mitigate injury risk. There is a substantial amount of scientific literature which investigates the ability of computer-vision and inertial sensor technologies to classify sport-specific movements. The coupling of automatic sport action labelling and athlete-monitoring data can significantly enhance athlete work-load monitoring. Two recent systematic reviews of the literature, pertinent to sport-specific movement classification, revealed that the majority of journal articles use athlete-dependent classification model training and evaluation methods. These methods can significantly enhance model classification performance, particularly with movements which have high inter-athlete technique variation. This is because it enables models to learn features distinctive to all athletes during training. This manuscript details the training and evaluation of supervised machine learning models to automatically classify running surface (athletics track, hard sand and soft sand) using features extracted from an upper-back inertial measurement unit sensor. Possible classification performance enhancement is demonstrated by comparing athlete independent and athlete dependent supervised machine learning methods. Using athlete dependent methods significantly increased the classification performance in terms of weighted average precision, recall, F1-score and accuracy (p < 0.05).