Towards Out-of-Lab Anterior Cruciate Ligament Injury Prevention and Rehabilitation Assessment: A Review of Portable Sensing Approaches DOI Creative Commons
Tian Tan, Anthony A. Gatti, Bingfei Fan

et al.

medRxiv (Cold Spring Harbor Laboratory), Journal Year: 2022, Volume and Issue: unknown

Published: Oct. 21, 2022

Abstract Anterior cruciate ligament (ACL) injury and ACL reconstruction (ACLR) surgery are common. Many ACL-injured subjects develop osteoarthritis within a decade of injury, major cause disability without cure. Laboratory-based biomechanical assessment can evaluate risk rehabilitation progress after ACLR; however, lab-based measurements expensive inaccessible to majority people. Portable sensors such as wearables cameras be deployed during sporting activities, in clinics, patient homes for assessment. Although many portable sensing approaches have demonstrated promising results various assessments related they not yet been widely adopted tools prevention training, evaluation reconstructions, return-to-sport decision making. The purpose this review is summarize research on out-of-lab applied ACLR offer our perspectives new opportunities future development. We identified 49 original articles ACL-related assessment; the most common modalities were inertial measurement units (IMUs), depth cameras, RGB cameras. studies combined with direct feature extraction, physics-based modeling, or machine learning estimate range parameters (e.g., knee kinematics kinetics) jump-landing tasks, cutting, squats, gait. reviewed depict proof-of-concept methods potential clinical applications including screening, By synthesizing these results, we describe important that exist using sophisticated modeling techniques enable more accurate along standardization data collection creation large benchmark datasets. If successful, advances will widespread use portable-sensing identify factors, mitigate high-risk movements prior optimize paradigms.

Language: Английский

Real-Time Multi-Person Identification and Tracking via HPE and IMU Data Fusion DOI
Mirco De Marchi, Cristian Turetta, Graziano Pravadelli

et al.

Published: March 25, 2024

Language: Английский

Citations

1
An IMU-Based Ground Reaction Force Estimation Method and Its Application in Walking Balance Assessment DOI Creative Commons
Xiangzhi Liu, Xiangliang Zhang, Bin Zhang

et al.

IEEE Transactions on Neural Systems and Rehabilitation Engineering, Journal Year: 2023, Volume and Issue: 32, P. 223 - 232

Published: Dec. 28, 2023

Walking is one of the most common daily movements human body. Therefore, quantitative evaluation walking has been commonly used to assist doctors in grasping disease degree and rehabilitation process patients clinic. Compared with kinematic characteristics, ground reaction force (GRF) during can directly reflect dynamic characteristics walking. It further help understand muscle recovery joint coordination patients. This paper proposes a GRF estimation method based on elastic elements Newton-Euler equation hybrid driving method. existing research, innovations are as follows. i) The hardware system consists only two inertial measurement units (IMUs) placed shanks. acquisition overall motion realized through simplified four-link model thigh prediction ii) was validated not 10 healthy subjects but also 11 Parkinson's stroke normalized mean absolute errors (NMAEs) 5.95%±1.32%, 6.09%±2.00%, 5.87%±1.59%. iii) balance assessment acquired data estimated GRF. evaluates ability fall risk at four key time points for all recruited. Because low-cost system, ease use, low interference environmental constraints, high accuracy, proposed automatic have broad clinical value.

Language: Английский

Citations

3
Estimation of Lower Limb Joint Angles and Joint Moments during Different Locomotive Activities Using the Inertial Measurement Units and a Hybrid Deep Learning Model DOI Creative Commons
Fanjie Wang, Wenqi Liang, H. M. Rehan Afzal

et al.

Sensors, Journal Year: 2023, Volume and Issue: 23(22), P. 9039 - 9039

Published: Nov. 8, 2023

Using inertial measurement units (IMUs) to estimate lower limb joint kinematics and kinetics can provide valuable information for disease diagnosis rehabilitation assessment. To gait parameters using IMUs, model-based filtering approaches have been proposed, such as the Kalman filter complementary filter. However, these methods require special calibration alignment of IMUs. The development deep learning algorithms has facilitated application IMUs in biomechanics it does not particular procedures use. hip/knee/ankle angles moments sagittal plane, a subject-independent temporal convolutional neural network-bidirectional long short-term memory network (TCN-BiLSTM) model was proposed three A public benchmark dataset containing most representative locomotive activities daily life used train evaluate TCN-BiLSTM model. mean Pearson correlation coefficient estimated by reached 0.92 0.87, respectively. This indicates that effectively multiple scenarios, demonstrating its potential clinical scenarios.

Language: Английский

Citations

2
Stride Width Estimation in Individuals with and without Neurodegenerative Disease via a Novel Data-Augmentation Deep Learning Model and Minimal Wearable Inertial Sensors DOI Creative Commons
Hong Wang,

Zakir Ullah,

Eran Gazit

et al.

Published: June 5, 2024

Stride width is vital for gait stability, postural balance control, and fall risk reduction. However, estimating stride typically requires either grounded cameras or a full kinematic body suit of wearable inertial measurement units (IMUs), both which are often too expensive time-consuming clinical application. We thus propose novel data-augmented deep learning model in individuals with without neurodegenerative disease using minimal set IMUs. Twelve patients neurodegenerative, clinically diagnosed Spinocerebellar ataxia type 3 (SCA3) performed over ground walking trials, seventeen healthy treadmill trials at various speeds modifications while wearing IMUs on each shank the pelvis. Results demonstrated mean absolute errors 3.3±0.7cm 2.9±0.5cm groups, respectively, were below important difference 6.0cm. variability 1.5cm 0.8cm respectively. Data augmentation significantly improved accuracy performance group, likely because they exhibited larger variations kinematics as compared subjects. These results could enable meaningful accurate portable monitoring disease, potentially enhancing rehabilitative training, assessment, dynamic control real-life settings.

Language: Английский

Citations

0
Kalman filter-based deep fused architecture for knee angle estimation DOI

Satheesh Kumar E,

Shyam Sundar

Multimedia Tools and Applications, Journal Year: 2024, Volume and Issue: unknown

Published: June 19, 2024

Language: Английский

Citations

0
Continuous mobile measurement of camptocormia angle using four accelerometers DOI Creative Commons
K. Naderi Beni,

Kathleen Knutzen,

J. P. Kuhtz-Buschbeck

et al.

Medical & Biological Engineering & Computing, Journal Year: 2024, Volume and Issue: 62(12), P. 3637 - 3652

Published: June 27, 2024

Abstract Camptocormia, a severe flexion deformity of the spine, presents challenges in monitoring its progression outside laboratory settings. This study introduces customized method utilizing four inertial measurement unit (IMU) sensors for continuous recording camptocormia angle (CA), incorporating both consensual malleolus and perpendicular assessment methods. The setup is wearable mobile allows measurements environment. practicality measuring CA across various activities evaluated mimicked Parkinson disease posture. Multiple are performed by healthy volunteer. Measurements compared against camera-based reference system. Results show an overall root mean squared error (RMSE) 4.13° 2.71° method. Furthermore, patient-specific calibration during standing still with forward lean activity significantly reduced RMSE to 2.45° 1.68° respectively. novel approach setting. proposed system suitable as tool first time implements IMU. It holds promise effectively at home.

Language: Английский

Citations

0
Towards Unstructured Unlabeled Optical Mocap: A Video Helps! DOI
Nicholas Milef, John Keyser, Shu Kong

et al.

Published: July 12, 2024

Optical motion capture (mocap) requires accurately reconstructing the human body from retroreflective markers, including pose and shape. In a typical mocap setting, marker labeling is an important but tedious error-prone step. Previous work has shown that can be automated by using structured template defining specific placements, this places additional recording constraints. We propose to relax these constraints solve for Unstructured Unlabeled (UUO) mocap. Compared setting either labels markers or them w.r.t layout, in UUO placed anywhere on even one limb (e.g., right leg biomechanics research), hence it of more practical significance. It also challenging. To mocap, we exploit monocular video captured single RGB camera, which does not require camera calibration. On video, run off-the-shelf method reconstruct track individual, giving strong visual priors With both optimization pipeline towards identification, labeling, estimation, reconstruction. Our technical novelties include multiple hypothesis testing optimize global orientation, localization marker-part matching better surface. conduct extensive experiments quantitatively compare our against state-of-the-art approaches, marker-only video-only body/shape Experiments demonstrate resoundingly outperforms existing methods three established benchmark datasets full-body partial-body

Language: Английский

Citations

0
Ankle Moment Estimation Based on A Novel Distributed Plantar Pressure Sensing System DOI

Mingyu Du,

Bowen Lv, Bingfei Fan

et al.

IEEE Journal of Biomedical and Health Informatics, Journal Year: 2024, Volume and Issue: 28(11), P. 6548 - 6556

Published: Aug. 16, 2024

Ankle moment plays an important role in human gait analysis, patients' rehabilitation process monitoring, and the human-machine interaction control of exoskeleton robots. However, current ankle estimation methods mainly rely on inverse dynamics (ID) based optical motion capture system (OMC) force plate. These fixed instruments laboratory, which are difficult to be applied To solve this problem, paper developed a new distributed plantar pressure proposed flexion method using system. We integrated eight sensors each insole collect data key area foot then used train four neural networks obtain moment. The performance models was evaluated normalized root mean square error (NRMSE) cross-correlation coefficient (ρ). During experiments, subjects were recruited for overground walking tests, OMC plate as gold standard. results indicate that Genetic algorithm - Gated recurrent unit (GA-GRU) best model achieved highest accuracy generalized (NRMSE = 7.23%, ρ 0.85) compared with other models. designed novel could serve joint approach wearable robot state monitoring.

Language: Английский

Citations

0
AddBiomechanics Dataset: Capturing the Physics of Human Motion at Scale DOI
Keenon Werling, Janelle M. Kaneda, Tian Tan

et al.

Lecture notes in computer science, Journal Year: 2024, Volume and Issue: unknown, P. 490 - 508

Published: Nov. 7, 2024

Language: Английский

Citations

0
Subject-Independent Ankle Joint Power Estimation with Two IMUs During Flat and Inclined Walking DOI
Hong Wang, Dongxuan Li,

Kairan Liang

et al.

Published: Oct. 9, 2023

Assessing ankle joint power during real-life scenarios is crucial for analyzing human push-off and detecting abnormal gait patterns. However, traditional monitoring methods require expensive professional equipment, limiting their use to laboratories. To address this limitation, we propose a portable robust two-stage approach that estimates using two inertial measurement units (IMU) sensors placed on the shank foot, respectively. Our subject-independent CNN model accurately assessed flat inclined walking across 28 speeds 6 ramp inclines. This solution facilitates assessment outside of laboratories could serve as foundation enable abnormality evaluation in patients hospitals, clinics, home-based settings.

Language: Английский

Citations

0