Deep learning based ground reaction force estimation for stair walking using kinematic data

Liu, Dongwei; He, Min; Hou, Meijin; Ma, Ye

doi:10.1016/j.measurement.2022.111344

Cited by 6 publications

(6 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A 1.8-m walkway and a customized eight-step staircase ( Liu et al, 2022 ) (height: 20 cm, tread depth: 30 cm, see Figure 2 ) with handrails on both sides were used to mimic stair transition scenarios in daily lives. At the top of the staircase, a 1-m long and 2.5-m wide platform with a fence around the platform was equipped for security.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Gait strategies for individuals with knee osteoarthritis when transitioning between floor and stair walking

et al. 2023

Self Cite

View full text Add to dashboard Cite

Objective: Individuals with knee osteoarthritis are at higher risk for falls during transitions between floor and stair walking due to their reduced lower extremity function. However, the adjust gait characteristics of knee osteoarthritis subjects for dealing with stair transitions have not been explored. We aimed to investigate gait strategies in individuals with knee osteoarthritis compared to age-matched individuals without knee OA during the transition between walking on floor and stairs.Method: Gait of 30 individuals with knee osteoarthritis and 30 individuals without knee osteoarthritis during floor-to-stair and stair-to-floor walking transitions were accessed using a 3D motion capture system. Foot-tread clearance, and angles of lower extremity joints and trunk at Foot-tread clearance timepoint were analyzed by using biomechanical software (Visual 3D).Results: Compared with asymptomatic control group, the knee osteoarthritis group demonstrated no difference in vertical Foot-tread clearance and horizontal Foot-tread clearance during stair transitions. However, ankle dorsiflexion (p = 0.01) decreased, hip flexion (p = 0.02) and trunk lateral tilt (p = 0.02) increased in knee osteoarthritis group during the stair ascent transition. Moreover, trunk lateral tilt to the support side (p = 0.003) and pelvic rotation (p = 0.03) increased, while hip abduction of the swing leg (p = 0.03) decreased during the stair descent transition in individuals with knee osteoarthritis.Conclusion: Increased trunk lateral tilt and altered angle of hip may be a strategy utilized by individuals with knee OA to increase the foot clearance ability and compensate for the disease-related loss of lower extremity strength, range of motion, and balance. However, compensatory manifestations, such as the increased lateral tilt of the trunk and movement of the gravity center may enhance the risk of falls and result in more abnormal knee joint loading.

show abstract

Section: Methodsmentioning

confidence: 99%

“…Following the administration of the clinical tests, 75 reflective markers, according to the calibrated anatomical systems technique protocol ( Cappozzo et al, 1995 ; Liu et al, 2022 ), were attached to the anatomical landmarks of each subject ( Figure 3 ; Table.1 ). The four markers on the feet were secured to sports shoes.…”

Section: Methodsmentioning

confidence: 99%

Gait strategies for individuals with knee osteoarthritis when transitioning between floor and stair walking

et al. 2023

Self Cite

View full text Add to dashboard Cite

show abstract

“…It is worth mentioning that the authors in [76] created a human motion capture database enhanced with pressure insoles data, which can be used for both animation and simulation. Similarly, in [77], vertical, anterior-posterior, and mediallateral GRFs during stair walking were estimated by a Bidirectional LSTM (an LSTM that has both forward and backward connections exploiting both past and future data) based on joint kinematics. Ground reaction forces can be used for both biomechanical analyses and therefore clinical applications as well as used in eliminating motion artifacts during simulation/animation as stated in [77].…”

Section: Musculoskeletal Dynamics Estimationmentioning

confidence: 99%

“…No GRFs Gait (Sohane, 2020) [91] Offline No GRFs Squatting (Zhang J., 2022) [7] Offline No GRFs Walking and Wrist Motions (Mourot, 2022) [76] Offline No GRFs Walking, running, jumping, etc. (Liu, 2022) [77] Offline No GRFs Stair Walking (Zhang Z., 2022) [8] Offline No GRFs Various from H3.6m [92] (Ehsani, 2020) [83] Offline No GRFs Object Manipulation Motions (Lee, 2019) [78] Offline No GRFs Walking, Running, Jumping, Kicking, etc. a The studies in Musculoskeletal Dynamics Estimation can be classified as follows: (i) studies that do or do not make use of motiondependent variables such as GRFs, (ii) works that developed either real-time or offline frameworks and (iii) according to the targeted activity (e.g.…”

Section: Publicationmentioning

confidence: 99%

Machine Learning Approaches for 3D Motion Synthesis and Musculoskeletal Dynamics Estimation: A Survey

Loi,

Zacharaki,

Moustakas

2024

IEEE Trans. Visual. Comput. Graphics

View full text Add to dashboard Cite

The inference of 3D motion and dynamics of the human musculoskeletal system has traditionally been solved using physics-based methods that exploit physical parameters to provide realistic simulations. Yet, such methods suffer from computational complexity and reduced stability, hindering their use in computer graphics applications that require real-time performance. With the recent explosion of data capture (mocap, video) machine learning (ML) has started to become popular as it is able to create surrogate models harnessing the huge amount of data stemming from various sources, minimizing computational time (instead of resource usage), and most importantly, approximate real-time solutions. The main purpose of this paper is to provide a review and classification of the most recent works regarding motion prediction, motion synthesis as well as musculoskeletal dynamics estimation problems using ML techniques, in order to offer sufficient insight into the state-of-the-art and draw new research directions. While the study of motion may appear distinct to musculoskeletal dynamics, these application domains provide jointly the link for more natural computer graphics character animation, since ML-based musculoskeletal dynamics estimation enables modeling of more long-term, temporally evolving, ergonomic effects, while offering automated and fast solutions. Overall, our review offers an in-depth presentation and classification of ML applications in human motion analysis, unlike previous survey articles focusing on specific aspects of motion prediction.

show abstract

“…The BiLSTM model plays a pivotal role in medical image feature extraction and sports injury prediction (Liu et al, 2022 ). Through continuous optimization of the model's structure, training strategy, and resolution of gradient problems, and leveraging domain expertise, the BiLSTM model is expected to become a potent tool for medical image and biomechanical data analysis, ensuring more accurate protection of athletes' health and performance.…”

Section: Related Workmentioning

confidence: 99%

Leveraging natural cognitive systems in conjunction with ResNet50-BiGRU model and attention mechanism for enhanced medical image analysis and sports injury prediction

Xiao,

Zhu,

Jiang

et al. 2023

Front. Neurosci.

View full text Add to dashboard Cite

IntroductionIn this study, we explore the potential benefits of integrating natural cognitive systems (medical professionals' expertise) and artificial cognitive systems (deep learning models) in the realms of medical image analysis and sports injury prediction. We focus on analyzing medical images of athletes to gain valuable insights into their health status.MethodsTo synergize the strengths of both natural and artificial cognitive systems, we employ the ResNet50-BiGRU model and introduce an attention mechanism. Our goal is to enhance the performance of medical image feature extraction and motion injury prediction. This integrated approach aims to achieve precise identification of anomalies in medical images, particularly related to muscle or bone damage.ResultsWe evaluate the effectiveness of our method on four medical image datasets, specifically pertaining to skeletal and muscle injuries. We use performance indicators such as Peak Signal-to-Noise Ratio and Structural Similarity Index, confirming the robustness of our approach in sports injury analysis.DiscussionOur research contributes significantly by providing an effective deep learning-driven method that harnesses both natural and artificial cognitive systems. By combining human expertise with advanced machine learning techniques, we offer a comprehensive understanding of athletes' health status. This approach holds potential implications for enhancing sports injury prevention, improving diagnostic accuracy, and tailoring personalized treatment plans for athletes, ultimately promoting better overall health and performance outcomes. Despite advancements in medical image analysis and sports injury prediction, existing systems often struggle to identify subtle anomalies and provide precise injury risk assessments, underscoring the necessity of a more integrated and comprehensive approach.

show abstract

Deep learning based ground reaction force estimation for stair walking using kinematic data

Cited by 6 publications

References 48 publications

Gait strategies for individuals with knee osteoarthritis when transitioning between floor and stair walking

Gait strategies for individuals with knee osteoarthritis when transitioning between floor and stair walking

Machine Learning Approaches for 3D Motion Synthesis and Musculoskeletal Dynamics Estimation: A Survey

Leveraging natural cognitive systems in conjunction with ResNet50-BiGRU model and attention mechanism for enhanced medical image analysis and sports injury prediction

Contact Info

Product

Resources

About