An Advanced Attention-Enhanced Hybrid Deep Learning Model for WBSNs’ Gait Pattern Recognition

Background: The deep learning techniques have been attracted increasing attention on wireless body sensor networks (WBSNs) gait pattern recognition that has a great contribution to monitoring gait change in clinical application. However, in existing studies, there are some challenging issues such as low generalization performance and no potential interpretation for gait variability. It is necessary to search for the advanced deep learning models to resolve these issues. Method: A public WARD database including acceleration and gyroscope data acquired from each subject wearing five sensors was selected, and the gait with different combination of on-body multi-sensors is considered as a WBSNs’ gait pattern. An advanced attention-enhanced hybrid deep learning model of DCNN and LSTM for WBSNs’ gait pattern recognition was proposed. In our proposed technique, the combination model of DCNN with LSTM is firstly to discover the spatial-temporary gait correlation features. And then the attention mechanism is introduced to exploit the more valuable intrinsic nonlinear dynamic correlation gait characteristics associated with gait variability hidden in spatial-temporary gait space obtained. This significantly contributes to enhancing the generalization performance and taking insight on gait variability in a certain anatomical region. Results: The ten gait patterns are randomly selected from WARD database to evaluate the feasibility of our proposed method. Our experiments demonstrated the superior generalization ability of our method to some models such as CNN-LSTM, DCNN-LSTM. Our proposed model could classify ten gait patterns with the highest accuracy and F1-score of 91.48% and 91.46%, respectively. Moreover, we also found that the classification performance of a certain gait pattern was almost same best when the combinations of three or five on-body sensors were employed respectively, suggesting that our method possibly take insight on gait variability in a certain anatomical region. Conclusion: Our proposed technique could feasibly discover the more intrinsic nonlinear dynamic correlation gait characteristics associated with gait variability from on-body multi-sensors gait data, which greatly contributed to best generalization performance and potential clinical interpretation. Our proposed technique would hopefully become a powerful tool of monitoring gait change in clinical application.

A Novel Mouse Model to Study Fracture Healing of Tibia

Background Delayed union of most tibial fractures due to their special anatomical structures.So an effective animal model is very important to study the mechanism and method of fracture healing.However, due to the small tibia of mice, the operation is difficult, and the surgical model requires high surgical skills. The construction of the fixation model of intramedullary nail for this fracture has improved and simplified the traditional fixation model of intramedullary nail, which not only achieves the purpose of constructing the fracture model, but also makes it more simple and effective.Therefore, the aim of the current study was to develop a new mouse model to study fracture healing of tibia. Methods We chose a combination between an open osteotomy and intramedullary stabilization. The 22G needle was inserted into the fracture end in a closed manner by using an open approach for osteotomy at the middle and lower 1/3 level of the tibia.Fractured tibia were analyzed using microcomputed tomography and histology at days 7,14,21and 28after surgery. All animals displayed normal limb loading and a physio-logical gait pattern within the first three days after fracture. No animals were lost due to surgery or anesthesia. Results X-ray confirmed that the fracture types obtained by the fracture modeling method were transverse fractures. X-ray, Micro-CT, immunohistochemistry, histological staining and Real-time PCR showed that the fracture healing of mice was typical endochondral ossification, with high repeatability. Conclusion The mouse tibial fracture model established by intramedullary nailing is safe, rapid and simple. Its fracture healing is a typical intrachondral ossification with high repeatability, which can be better used for the study of molecular mechanism and clinical transformation of fracture healing and bone metabolism.

A Review of Existing Transtibial Bionic Prosthesis: Mechanical Design, Actuators and Power Transmission

Transtibial and transfemoral amputations are the most common amputations in the world, loss of lower extremity result in impaired function extremities and also body balance. A prosthesis is a medical device designed to replace a specific body part to restore function to a body part lost due to an accident or disease. Most doctors strongly recommend the use of a prosthesis so that patients can return to normal activities after undergoing an amputation. Besides functioning to support beauty, the use of prostheses is also to restore the quality of life of prosthetic users, the issue of metabolic energy consumption when walking is also very important in designing transtibial bionic prosthesis because it involves the comfort of the user transtibial prosthesis. Most of the existing transtibial prosthesis products in Indonesia are conventional passive transtibial foot products, and passive prosthesis users show a limp or asymmetrical gait pattern so that conventional passive prosthesis users experience discomfort when walking in the form of pain in the amputated leg and normal foot, which can cause secondary musculoskeletal injuries such as joint disorders. Passive prostheses cannot generate propulsive force during push-off phase (terminal stance and preswing) of the human gait cycle. The use of passive prostheses can also consume 20-30% more metabolic energy while walking so that it can cause fatigue for the user. Transtibial bionic prosthesis research is growing, transtibial bionic prosthesis can overcome the weakness of passive prosthesis because it can produce push-off during gait cycle and several researchers have shown that bionic prostheses are capable of mimicking the human gait, as well as improve the performance in a more natural gait and normal walking. This study aims to study the existing transtibial bionic prosthesis by comparing between 6 existing designs of powered ankle or transtibial bionic prosthesis that have been published in several publications. The discussion focuses on the design and mechanical systems, actuators related to the selection of motors and drive mechanisms as well as power transmission from actuators to moving components.

Efficiency and Stability of Step-To Gait in Slow Walking

As humans, we constantly change our movement strategies to adapt to changes in physical functions and the external environment. We have to walk very slowly in situations with a high risk of falling, such as walking on slippery ice, carrying an overflowing cup of water, or muscle weakness owing to aging or motor deficit. However, previous studies have shown that a normal gait pattern at low speeds results in reduced efficiency and stability in comparison with those at a normal speed. Another possible strategy is to change the gait pattern from normal to step-to gait, in which the other foot is aligned with the first swing foot. However, the efficiency and stability of the step-to gait pattern at low speeds have not been investigated yet. Therefore, in this study, we compared the efficiency and stability of the normal and step-to gait patterns at intermediate, low, and very low speeds. Eleven healthy participants were asked to walk with a normal gait and step-to gait on a treadmill at five different speeds (i.e., 10, 20, 30, 40, and 60 m/min), ranging from very low to normal walking speed. The efficiency parameters (percent recovery and walk ratio) and stability parameters (center of mass lateral displacement) were analyzed from the motion capture data and then compared for the two gait patterns. The results suggested that step-to gait had a more efficient gait pattern at very low speeds of 10–30 m/min, with a larger percent recovery, and was more stable at 10–60 m/min in comparison with a normal gait. However, the efficiency of the normal gait was better than that of the step-to gait pattern at 60 m/min. Therefore, step-to gait is effective in improving gait efficiency and stability when faced with situations that force us to walk slowly or hinder quick walking because of muscle weakness owing to aging or motor deficit along with a high risk of falling.

EVALUATION OF CHONDROPROTECTIVE EFFECTS OF HYALURONIC ACID IN A RAT MODEL OF OSTEOARTHRITIS

Objective: To evaluate the chondroprotective effects of hyaluronic acid in a rat model of osteoarthritis. Study Design: Laboratory based experimental study. Place and Duration of Study: Department of Pharmacology, Army Medical College, Rawalpindi, in alliance with National Institute of health, Islamabad and Department of Pathology, Army Medical College Rawalpindi, from Apr to Jun 2019. Methodology: Sixteen (16) rats of Sprague Dawley breed were procured in this study. Osteoarthritis was induced in right knee joint of rats by surgical resection of medial meniscus and anterior cruciate ligament. They were allocated into two (02) groups with eight (08) rats in both groups. Group-I was control group that was treated with 0.2 ml intra articular saline once weekly for four weeks. While group-II was treatment group that was intra particularly administered with 0.2ml hyaluronic acid once weekly for four weeks. One week after the last dosage, gait pattern of the animals was scored. Then animals were sacrificed and a part of proximal tibia was obtained for histopathologic analysis. Results: Mean gait score of control group and treatment group was 3.25 ± 0.707 and 1.00 ± 0.756 respectively with a statistically significant p-value of <0.001, while mean histopathological Modified Mankin score of control and treatment group was 11.5 ± 1.195 and 5.50 ± 1.195 respectively with a significant p-value of <0.001. Conclusion: Intra articular viscosupplementation of hyaluronic acid in rat model of osteoarthritis resulted in improved gait pattern and histopathology.

The Impact of Different Periods of Walking Experience on Kinematic Gait Parameters in Toddlers

This study aimed to analyse the kinematic differences in gait between three groups of toddlers who differed in their weeks of independent walking (IW) experience, but not in anthropometrical characteristics, to determine the relationship between walking experience without the side effect of morphological differences on gait parameters. Twenty-six toddlers participated in this study. Depending on the week of their IW, toddlers were divided into three groups: Group 1 (1–5 weeks of IW), Group 2 (6–10 weeks of IW), and Group 3 (11–15 weeks of IW). Each toddler walked barefooted over a 2-m long pathway, and 3D kinematic data were obtained. A decrease in the upper limb position, hip flexion, and step width, i.e., changes towards the adult gait pattern, were observed in Group 3. Less experienced walkers exhibited a wider step width despite no statistically significant difference in body mass and height between groups. Results of this study show no statistically significant difference in step length between groups, suggesting that step length is more related to height than to the walking experience. The increased step length in more experienced walkers reported in previous studies may therefore be a result of different heights and not walking experience.

Human Gait Data Augmentation and Trajectory Prediction for Lower-Limb Rehabilitation Robot Control Using GANs and Attention Mechanism

To date, several alterations in the gait pattern can be treated through rehabilitative approaches and robot assisted therapy (RAT). Gait data and gait trajectories are essential in specific exoskeleton control strategies. Nevertheless, the scarcity of human gait data due to the high cost of data collection or privacy concerns can hinder the performance of controllers or models. This paper thus first creates a GANs-based (Generative Adversarial Networks) data augmentation method to generate synthetic human gait data while still retaining the dynamics of the real gait data. Then, both the real collected and the synthesized gait data are fed to our constructed two-stage attention model for gait trajectories prediction. The real human gait data are collected with the five healthy subjects recruited from an optical motion capture platform. Experimental results indicate that the created GANs-based data augmentation model can synthesize realistic-looking multi-dimensional human gait data. Also, the two-stage attention model performs better compared with the LSTM model; the attention mechanism shows a higher capacity of learning dependencies between the historical gait data to accurately predict the current values of the hip joint angles and knee joint angles in the gait trajectory. The predicted gait trajectories depending on the historical gait data can be further used for gait trajectory tracking strategies.

A model-free deep reinforcement learning approach for control of exoskeleton gait patterns

Lower-body exoskeleton control that adapts to users and provides assistance-as-needed can increase user participation and motor learning and allow for more effective gait rehabilitation. Adaptive model-based control methods have previously been developed to consider a user’s interaction with an exoskeleton; however, the predefined dynamics models required are challenging to define accurately, due to the complex dynamics and nonlinearities of the human-exoskeleton interaction. Model-free deep reinforcement learning (DRL) approaches can provide accurate and robust control in robotics applications and have shown potential for lower-body exoskeletons. In this paper, we present a new model-free DRL method for end-to-end learning of desired gait patterns for over-ground gait rehabilitation with an exoskeleton. This control technique is the first to accurately track any gait pattern desired in physiotherapy without requiring a predefined dynamics model and is robust to varying post-stroke individuals’ baseline gait patterns and their interactions and perturbations. Simulated experiments of an exoskeleton paired to a musculoskeletal model show that the DRL method is robust to different post-stroke users and is able to accurately track desired gait pattern trajectories both seen and unseen in training.

The effect of constraining mediolateral ankle moments and foot placement on the use of the counter-rotation mechanism during walking

Stability during walking can be maintained by shifts of the Center of Pressure through modulation of foot placement and ankle moments (CoP-mechanism). An additional mechanism to stabilize gait, is the counter-rotation mechanism i.e. changing the angular momentum of segments around the Center of Mass (CoM) to change the direction of the ground reaction force. It is unknown if and how humans use the counter-rotation mechanism to control the CoM during walking and how this interacts with the CoP-mechanism. Thirteen healthy adults walked on a treadmill, while full-body kinematic and force plate data were obtained. The contributions of the CoP and the counter-rotation mechanisms to control the CoM were calculated during steady-state walking, walking on LesSchuh, i.e. constraining mediolateral CoP shifts underneath the stance foot and walking on LesSchuh at 50% of normal step width, constraining both foot placement and ankle mechanisms (LesSchuh50%). A decreased magnitude of within-stride control by the CoP-mechanism was compensated for by an increased magnitude of within-stride control by the counter-rotation mechanism during LesSchuh50% compared to steady-state walking. This suggests that the counter-rotation mechanism is used to stabilize gait when needed. However, the mean contribution of the counter-rotation mechanism over strides did not increase during LesSchuh50% compared to steady-state walking. The CoP-mechanism was the main contributor to the total CoM acceleration. The use of the counter-rotation mechanism may be limited because angular accelerations ultimately need to be reversed and because of interference with other task constraints, such as head stabilization and preventing interference with the gait pattern.