Context-informed incremental learning improves both the performance and resilience of myoelectric control.

Despite its rich history of success in controlling powered prostheses and emerging commercial interests in ubiquitous computing, myoelectric control continues to suffer from a lack of robustness. In particular, EMG-based systems often degrade over prolonged use resulting in tedious recalibration sessions, user frustration, and device abandonment. Unsupervised adaptation is one proposed solution that updates a model's parameters over time based on its own predictions during real-time use to maintain robustness without requiring additional user input or dedicated recalibration. However, these strategies can actually accelerate performance deterioration when they begin to classify (and thus adapt) incorrectly, defeating their own purpose. To overcome these limitations, we propose a novel adaptive learning strategy, Context-Informed Incremental Learning (CIIL), that leverages in situ context to better inform the prediction of pseudo-labels. In this work, we evaluate these CIIL strategies in an online target acquisition task for two use cases: (1) when there is a lack of training data and (2) when a drastic and enduring alteration in the input space has occurred. A total of 32 participants were evaluated across the two experiments. The results show that the CIIL strategies significantly outperform the current state-of-the-art unsupervised high-confidence adaptation and outperform models trained with the conventional screen-guided training approach, even after a 45-degree electrode shift (p < 0.05). Consequently, CIIL has substantial implications for the future of myoelectric control, potentially reducing the training burden while bolstering model robustness, and leading to improved real-time control.

Botulinum toxin and hyperhidrosis of the amputation stump in war amputees.

INTRODUCTION: Stump hyperhidrosis is a common condition after lower limb amputation. It affects the prosthesis use, and the quality of life of patients. Several case reports tried to prove benefit of using Botulinum toxin in its treatment. AIM: This study was to conduct a larger workforce clinical trial and to demonstrate benefits of botulinum toxin injection in the treatment of stump hyperhidrosis. METHODS: A prospective study was conducted. War amputees who complained of annoying excessive sweating of the stump were included. They received intradermal injection of botulinum toxin A in the residual limb area in contact with prosthetic socket. Abundance of sweating and degree of functional discomfort associated with it were assessed before, after 3 weeks, 6 and 12 months. RESULTS: Seventeen male patients, followed for post-traumatic limb amputation were included in the study. Discomfort and bothersome in relation to Hyperhidrosis did decrease after treatment (p<0,001). Reported satisfaction after 3 weeks was 73,33%. Improvement of prothesis loosening up after 3 weeks was 72,5% [±15,6]. Mean injection-induced pain on the visual analogue scale was 5.17/10 (±1.58). The mean interval after the onset of improvement was 5.13 days [min:3, max:8]. The mean time of improvement was 10.4 months after the injection [min:6, max:12]. No major adverse events were reported following treatment. CONCLUSION: Intradermal injections of botulinum toxin in the symptomatic treatment of stump hyperhidrosis are effective and have few adverse effects. It improves the quality of life of our patients thanks to a better tolerance of the prosthesis.

Understanding the influence of confounding factors in myoelectric control for discrete gesture recognition.

Discrete myoelectric control-based gesture recognition has recently gained interest as a possible input modality for many emerging ubiquitous computing applications. Unlike the continuous control commonly employed in powered prostheses, discrete systems seek to recognize the dynamic sequences associated with gestures to generate event-based inputs. More akin to those used in general-purpose human-computer interaction, these could include, for example, a flick of the wrist to dismiss a phone call or a double tap of the index finger and thumb to silence an alarm. Moelectric control systems have been shown to achieve near-perfect classification accuracy, but in highly constrained offline settings. Real-world, online systems are subject to 'confounding factors' (i.e. factors that hinder the real-world robustness of myoelectric control that are not accounted for during typical offline analyses), which inevitably degrade system performance, limiting their practical use. Although these factors have been widely studied in continuous prosthesis control, there has been little exploration of their impacts on discrete myoelectric control systems for emerging applications and use cases. Correspondingly, this work examines, for the first time, three confounding factors and their effect on the robustness of discrete myoelectric control: (1)limb position variability, (2)cross-day use, and a newly identified confound faced by discrete systems (3)gesture elicitation speed. Results from four different discrete myoelectric control architectures: (1) Majority Vote LDA, (2) Dynamic Time Warping, (3) an LSTM network trained with Cross Entropy, and (4) an LSTM network trained with Contrastive Learning, show that classification accuracy is significantly degraded (p<0.05) as a result of each of these confounds. This work establishes that confounding factors are a critical barrier that must be addressed to enable the real-world adoption of discrete myoelectric control for robust and reliable gesture recognition.

Evaluation of Prosthesis Satisfaction, Psychological Status, and Quality of Life in Lower-Extremity Amputee Patients: A Pilot Study.

BACKGROUND: There are various factors affecting the use of prostheses. This study was aimed to examine satisfaction, psychological state, quality of life, and the factors affecting these in individuals who use prostheses because of lower-extremity amputation. METHODS: Sixty-three patients were included in this study. Demographic data and features related amputation and prosthesis were recorded. Quality of life was evaluated with the Nottingham Health Profile (NHP), anxiety and depression levels were evaluated with the Hospital Anxiety Depression Scale (HADS), body image was evaluated by the Amputee Body Image Scale (ABIS), prosthesis satisfaction was evaluated with the Prosthesis Satisfaction Questionnaire (PSQ), and the relationship between them was examined. RESULTS: There was a positive correlation between all HADS scores, NHP-emotional reactions, NHP-sleep, NHP-social isolation, NHP-total, and ABIS (P < .05). A negative correlation was found between HADS-anxiety and PSQ results (r = -0.394, P = .003). A positive correlation was found between HADS-depression scores and NHP-pain, NHP-emotional reactions, NHP-social isolation, NHP-total, and ABIS (P < .05); and a negative correlation was found with PSQ questionnaire scores (r = -0.427, P = .001). There was a positive correlation between HADS-total scores and all parameters except NHP-energy level and ABIS (P < .05). A positive correlation was found between ABIS and all parameters except NHP-energy level (P < .05). A negative correlation was found between PSQ and NHP-social isolation, NHP-physical activity, and NHP-total scores (r = -0.312, P = 0.019; r = -0.312, P = .019; and r = -0.277, P = .039, respectively). The presence of residual extremity pain was found to be an effective factor on the psychological state (ß = 0.429, P = .001). The presence of residual limb pain and phantom pain were found to be effective factors on the prosthesis satisfaction (ß = -0.41, P = .001; and ß = -0.406, P = .001, respectively). The presence of residual extremity pain and anxiety level were found independent risk factors on the NHP (ß = -0.401, P = .006; and ß = -0.445, P = .006, respectively). CONCLUSIONS: Individuals using prostheses because of lower-extremity amputation should be examined in detail from various perspectives.

Biomechanical effects of adding an articulating toe joint to a passive foot prosthesis for incline and decline walking.

Walking on sloped surfaces is challenging for many lower limb prosthesis users, in part due to the limited ankle range of motion provided by typical prosthetic ankle-foot devices. Adding a toe joint could potentially benefit users by providing an additional degree of flexibility to adapt to sloped surfaces, but this remains untested. The objective of this study was to characterize the effect of a prosthesis with an articulating toe joint on the preferences and gait biomechanics of individuals with unilateral below-knee limb loss walking on slopes. Nine active prosthesis users walked on an instrumented treadmill at a +5° incline and -5° decline while wearing an experimental foot prosthesis in two configurations: a Flexible toe joint and a Locked-out toe joint. Three participants preferred the Flexible toe joint over the Locked-out toe joint for incline and decline walking. Eight of nine participants went on to participate in a biomechanical data collection. The Flexible toe joint decreased prosthesis Push-off work by 2 Joules during both incline (p = 0.008; g = -0.63) and decline (p = 0.008; g = -0.65) walking. During incline walking, prosthetic limb knee flexion at toe-off was 3° greater in the Flexible configuration compared to the Locked (p = 0.008; g = 0.42). Overall, these results indicate that adding a toe joint to a passive foot prosthesis has relatively small effects on joint kinematics and kinetics during sloped walking. This study is part of a larger body of work that also assessed the impact of a prosthetic toe joint for level and uneven terrain walking and stair ascent/descent. Collectively, toe joints do not appear to substantially or consistently alter lower limb mechanics for active unilateral below-knee prosthesis users. Our findings also demonstrate that user preference for passive prosthetic technology may be both subject-specific and task-specific. Future work could investigate the inter-individual preferences and potential benefits of a prosthetic toe joint for lower-mobility individuals.

An adaptive prosthetic socket for people with transtibial amputation.

It is essential that people with limb amputation maintain proper prosthetic socket fit to prevent injury. Monitoring and adjusting socket fit, for example by removing the prosthesis to add prosthetic socks, is burdensome and can adversely affect users' function and quality-of-life. This study presents results from take-home testing of a motor-driven adaptive socket that automatically adjusted socket size during walking. A socket fit metric was calculated from inductive sensor measurements of the distance between the elastomeric liner surrounding the residual limb and the socket's inner surface. A proportional-integral controller was implemented to adjust socket size. When tested on 12 participants with transtibial amputation, the controller was active a mean of 68% of the walking time. In general, participants who walked more than 20 min/day demonstrated greater activity, less doff time, and fewer manual socket size adjustments for the adaptive socket compared with a locked non-adjustable socket and a motor-driven socket that participants adjusted with a smartphone application. Nine of 12 participants reported that they would use a motor-driven adjustable socket if it were available as it would limit their socket fit issues. The size and weight of the adaptive socket were considered the most important variables to improve.

Lower-extremity inter-joint coordination variability in active individuals with transtibial amputation and healthy males during gait.

This study was aimed to compare the variability of inter-joint coordination in the lower-extremities during gait between active individuals with transtibial amputation (TTAs) and healthy individuals (HIs). Fifteen active male TTAs (age: 40.6 ± 16.24 years, height: 1.74 ± 0.09 m, and mass: 71.2 ± 8.87 kg) and HIs (age: 37.25 ± 13.11 years, height: 1.75 ± 0.06 m, and mass: 74 ± 8.75 kg) without gait disabilities voluntarily participated in the study. Participants walked along a level walkway covered with Vicon motion capture system, and their lower-extremity kinematics data were recorded during gait. The spatiotemporal gait parameters, lower-extremity joint range of motion (ROM), and their coordination and variability were calculated and averaged to report a single value for each parameter based on biomechanical symmetry assumption in the lower limbs of HIs. Additionally, these parameters were separately calculated and reported for the intact limb (IL) and the prosthesis limb (PL) in TTAs individuals. Finally, a comparison was made between the averaged values in HIs and those in the IL and PL of TTAs subjects. The results showed that the IL had a significantly lower stride length than that of the PL and averaged value in HIs, and the IL had a significantly lower knee ROM and greater stance-phase duration than that of HIs. Moreover, TTAs showed different coordination patterns in pelvis-to-hip, hip-to-knee, and hip-to-ankle couplings in some parts of the gait cycle. It concludes that the active TTAs with PLs walked with more flexion of the knee and hip, which may indicate a progressive walking strategy and the differences in coordination patterns suggest active TTA individuals used different neuromuscular control strategies to adapt to their amputation. Researchers can extend this work by investigating variations in these parameters across diverse patient populations, including different amputation etiologies and prosthetic designs. Moreover, Clinicians can use the findings to tailor rehabilitation programs for TTAs, emphasizing joint flexibility and coordination.

Excitation of natural spinal reflex loops in the sensory-motor control of hand prostheses.

Sensory feedback for prosthesis control is typically based on encoding sensory information in specific types of sensory stimuli that the users interpret to adjust the control of the prosthesis. However, in physiological conditions, the afferent feedback received from peripheral nerves is not only processed consciously but also modulates spinal reflex loops that contribute to the neural information driving muscles. Spinal pathways are relevant for sensory-motor integration, but they are commonly not leveraged for prosthesis control. We propose an approach to improve sensory-motor integration for prosthesis control based on modulating the excitability of spinal circuits through the vibration of tendons in a closed loop with muscle activity. We measured muscle signals in healthy participants and amputees during different motor tasks, and we closed the loop by applying vibration on tendons connected to the muscles, which modulated the excitability of motor neurons. The control signals to the prosthesis were thus the combination of voluntary control and additional spinal reflex inputs induced by tendon vibration. Results showed that closed-loop tendon vibration was able to modulate the neural drive to the muscles. When closed-loop tendon vibration was used, participants could achieve similar or better control performance in interfaces using muscle activation than without stimulation. Stimulation could even improve prosthetic grasping in amputees. Overall, our results indicate that closed-loop tendon vibration can integrate spinal reflex pathways in the myocontrol system and open the possibility of incorporating natural feedback loops in prosthesis control.

Long-term functional and clinical outcome of combined targeted muscle reinnervation and osseointegration for functional bionic reconstruction in transhumeral amputees: a case series.

OBJECTIVE: To describe and evaluate the combination of osseointegration and nerve transfers in 3 transhumeral amputees. DESIGN: Case series. PATIENTS: Three male patients with a unilateral traumatic transhumeral amputation. METHODS: Patients received a combination of osseointegration and targeted muscle reinnervation surgery. Rehabilitation included graded weight training, range of motion exercises, biofeedback, table-top prosthesis training, and controlling the actual device. The impairment in daily life, health-related quality of life, and pain before and after the intervention was evaluated in these patients. Their shoulder range of motion, prosthesis embodiment, and function were documented at a 2- to 5-year follow-up. RESULTS: All 3 patients attended rehabilitation and used their myoelectric prosthesis on a daily basis. Two patients had full shoulder range of motion with the prosthesis, while the other patient had 55° of abduction and 45° of anteversion. They became more independent in their daily life activities after the intervention and incorporated their prosthesis into their body scheme to a high extent. CONCLUSION: These results indicate that patients can benefit from the combined procedure. However, the patients' perspective, risks of the surgical procedures, and the relatively long rehabilitation procedure need to be incorporated in the decision-making.

Vision-aided grasp classification: design and evaluation of compact CNN for prosthetic hands.

Powered prosthetic hands capable of executing various grasp patterns are highly sought-after solutions for upper limb amputees. A crucial requirement for such prosthetic hands is the accurate identification of the intended grasp pattern and subsequent activation of the prosthetic digits accordingly. Vision-based grasp classification techniques offer improved coordination between amputees and prosthetic hands without physical contact. Deep learning methods, particularly Convolutional Neural Networks (CNNs), are utilized to process visual information for classification. The key challenge lies in developing a model that can effectively generalize across various object shapes and accurately classify grasp classes. To address this, a compact CNN model named GraspCNet is proposed, specifically designed for grasp classification in prosthetic hands. The use of separable convolutions reduces the computational burden, making it potentially suitable for real-time applications on embedded systems. The GraspCNet model is designed to learn and generalize from object shapes, allowing it to effectively classify unseen objects beyond those included in the training dataset. The proposed model was trained and tested using various standard object data sets. A cross-validation strategy has been adopted to perform better in seen and unseen object class scenarios. The average accuracy achieved was 82.22% and 75.48% in the case of seen, and unseen object classes respectively. In computer-based real-time experiments, the GraspCNet model achieved an accuracy of 69%. A comparative analysis with state-of-the-art techniques revealed that the proposed GraspCNet model outperformed most benchmark techniques and demonstrated comparable performance with the DcnnGrasp method. The compact nature of the GraspCNet model suggests its potential for integration with other sensing modalities in prosthetic hands.

Developing a Novel Prosthetic Hand with Wireless Wearable Sensor Technology Based on User Perspectives: A Pilot Study.

Myoelectric hands are beneficial tools in the daily activities of people with upper-limb deficiencies. Because traditional myoelectric hands rely on detecting muscle activity in residual limbs, they are not suitable for individuals with short stumps or paralyzed limbs. Therefore, we developed a novel electric prosthetic hand that functions without myoelectricity, utilizing wearable wireless sensor technology for control. As a preliminary evaluation, our prototype hand with wireless button sensors was compared with a conventional myoelectric hand (Ottobock). Ten healthy therapists were enrolled in this study. The hands were fixed to their forearms, myoelectric hand muscle activity sensors were attached to the wrist extensor and flexor muscles, and wireless button sensors for the prostheses were attached to each user's trunk. Clinical evaluations were performed using the Simple Test for Evaluating Hand Function and the Action Research Arm Test. The fatigue degree was evaluated using the modified Borg scale before and after the tests. While no statistically significant differences were observed between the two hands across the tests, the change in the Borg scale was notably smaller for our prosthetic hand (p = 0.045). Compared with the Ottobock hand, the proposed hand prosthesis has potential for widespread applications in people with upper-limb deficiencies.

Validity and Impact of Methods for Collecting Training Data for Myoelectric Prosthetic Control Algorithms.

Intuitive regression control of prostheses relies on training algorithms to correlate biological recordings to motor intent. The quality of the training dataset is critical to run-time regression performance, but accurately labeling intended hand kinematics after hand amputation is challenging. In this study, we quantified the accuracy and precision of labeling hand kinematics using two common training paradigms: 1) mimic training, where participants mimic predetermined motions of a prosthesis, and 2) mirror training, where participants mirror their contralateral intact hand during synchronized bilateral movements. We first explored this question in healthy non-amputee individuals where the ground-truth kinematics could be readily determined using motion capture. Kinematic data showed that mimic training fails to account for biomechanical coupling and temporal changes in hand posture. Additionally, mirror training exhibited significantly higher accuracy and precision in labeling hand kinematics. These findings suggest that the mirror training approach generates a more faithful, albeit more complex, dataset. Accordingly, mirror training resulted in significantly better offline regression performance when using a large amount of training data and a non-linear neural network. Next, we explored these different training paradigms online, with a cohort of unilateral transradial amputees actively controlling a prosthesis in real-time to complete a functional task. Overall, we found that mirror training resulted in significantly faster task completion speeds and similar subjective workload. These results demonstrate that mirror training can potentially provide more dexterous control through the utilization of task-specific, user-selected training data. Consequently, these findings serve as a valuable guide for the next generation of myoelectric and neuroprostheses leveraging machine learning to provide more dexterous and intuitive control.

Spike timing-based coding in neuromimetic tactile system enables dynamic object classification.

Rapid processing of tactile information is essential to human haptic exploration and dexterous object manipulation. Conventional electronic skins generate frames of tactile signals upon interaction with objects. Unfortunately, they are generally ill-suited for efficient coding of temporal information and rapid feature extraction. In this work, we report a neuromorphic tactile system that uses spike timing, especially the first-spike timing, to code dynamic tactile information about touch and grasp. This strategy enables the system to seamlessly code highly dynamic information with millisecond temporal resolution on par with the biological nervous system, yielding dynamic extraction of tactile features. Upon interaction with objects, the system rapidly classifies them in the initial phase of touch and grasp, thus paving the way to fast tactile feedback desired for neuro-robotics and neuro-prosthetics.

Development and Preliminary Evaluation of a Bimodal Foot Prosthesis for Walking and Running.

People often alternate between bouts of walking and running, for instance, when adults participate in recreational activities. Transitioning between activities can be challenging for prosthesis users because existing prosthetic feet are not well-suited for both tasks. Meanwhile, switching between prostheses for different tasks is often impractical. Collectively, these challenges can present barriers to physical activity participation for people with limb loss, which can negatively impact social or physical health. This work describes the development and evaluation of a passive bimodal prosthetic foot prototype with different configurations and stiffnesses for walking and running. Users rated the bimodal prosthesis higher for standing and walking compared to a running prosthesis (+2.3 for both tasks on a seven-point Likert scale). Users rated the bimodal prosthesis higher for running compared to a walking prosthesis (+1.7 and +0.5 for 2.0 and 2.5 m/s running, respectively). Changing from walking to running mode increased the device's stiffness by 23-84%, depending on the user's preference. Users could switch between bimodal prosthesis walking and running modes quickly (21.3 ± 12.0 s). Overall, the preliminary results were encouraging in terms of user satisfaction, stiffness change between modes, and mode-switching speed. These findings motivate future exploration of this bimodal prosthesis concept.

Simulation and experimental study on rope driven artificial hand and driven motor.

BACKGROUND: Prosthetic hands have the potential to replace human hands. Using prosthetic hands can help patients with hand loss to complete the necessary daily living actions. OBJECTIVE: This paper studies the design of a bionic, compact, low-cost, and lightweight 3D printing humanoid hand. The five fingers are underactuated, with a total of 9 degrees of freedom. METHODS: In the design of an underactuated hand, it is a basic element composed of an actuator, spring, rope, and guide system. A single actuator is providing power for five fingers. And the dynamic simulation is carried out to calculate the motion trajectory effect. RESULTS: In this paper, the driving structure of the ultrasonic motor was designed, and the structural size of the ultrasonic motor vibrator was determined by modal and transient simulation analysis, which replace the traditional brake, realize the lightweight design of prosthetic hand, improve the motion accuracy and optimize the driving performance of prosthetic hand. CONCLUSIONS: By replacing traditional actuators with new types of actuators, lightweight design of prosthetic hands can be achieved, improving motion accuracy and optimizing the driving performance of prosthetic hands.

A multifaceted suite of metrics for comparative myoelectric prosthesis controller research.

Upper limb robotic (myoelectric) prostheses are technologically advanced, but challenging to use. In response, substantial research is being done to develop person-specific prosthesis controllers that can predict a user's intended movements. Most studies that test and compare new controllers rely on simple assessment measures such as task scores (e.g., number of objects moved across a barrier) or duration-based measures (e.g., overall task completion time). These assessment measures, however, fail to capture valuable details about: the quality of device arm movements; whether these movements match users' intentions; the timing of specific wrist and hand control functions; and users' opinions regarding overall device reliability and controller training requirements. In this work, we present a comprehensive and novel suite of myoelectric prosthesis control evaluation metrics that better facilitates analysis of device movement details-spanning measures of task performance, control characteristics, and user experience. As a case example of their use and research viability, we applied these metrics in real-time control experimentation. Here, eight participants without upper limb impairment compared device control offered by a deep learning-based controller (recurrent convolutional neural network-based classification with transfer learning, or RCNN-TL) to that of a commonly used controller (linear discriminant analysis, or LDA). The participants wore a simulated prosthesis and performed complex functional tasks across multiple limb positions. Analysis resulting from our suite of metrics identified 16 instances of a user-facing problem known as the "limb position effect". We determined that RCNN-TL performed the same as or significantly better than LDA in four such problem instances. We also confirmed that transfer learning can minimize user training burden. Overall, this study contributes a multifaceted new suite of control evaluation metrics, along with a guide to their application, for use in research and testing of myoelectric controllers today, and potentially for use in broader rehabilitation technologies of the future.

Infection After Lower-Limb Osseointegration: A Single-Center Retrospective Evaluation of Pathogens, Management, and Outcomes.

PURPOSE: Osseointegration (OI) is a novel alternative to traditional socket-suspended prostheses for lower-limb amputees, eliminating the socket-skin interface and allowing for weight bearing directly on the skeletal system. However, the stoma through which the implant attaches to the external prosthesis creates an ingress route for bacteria, and infection rates as high as 66% have been reported. The aims of this study are to classify infection management and long-term outcomes in this patient population to maximize implant salvage. METHODS: An institutional review board-approved retrospective analysis was performed on all patients who underwent lower-limb OI at our institution between 2017 and 2022. Demographic, operative, and outcome data were collected for all patients. Patients were stratified by the presence and severity of infection. Chi-square and t tests were performed on categorical and continuous data, respectively, using an alpha of 0.05. RESULTS: One hundred two patients met our study criteria; 62 had transfemoral OI and 40 had transtibial OI. Patients were followed for 23.8 months on average (range, 3.5-63.7). Osteomyelitis was more likely than soft tissue infection to be polymicrobial in nature (71% vs 23%, P < 0.05). Infections at the stoma were mostly (96%) managed with oral antibiotics alone, whereas deeper soft tissue infections also required intravenous antibiotics (75%) or operative washout (19%). Osteomyelitis was managed with intravenous antibiotics and required operative attention; 5 (71%) underwent washout and 2 (29%) underwent explantation. Both implants were replaced an average of 3.5 months after explantation. There was no correlation between history of soft tissue infection and development of osteomyelitis (P > 0.05). The overall implant salvage rate after infection was 96%. CONCLUSIONS: This study describes our institution's experience managing infection after OI and soft tissue reconstruction. Although infections do occur, they are easily treatable and rarely require operative intervention. Explantation due to infection is rare and can be followed up with reimplantation, reaffirming that OI is a safe and effective treatment modality.

Towards efficient motor imagery interventions after lower-limb amputation.

BACKGROUND: The therapeutic benefits of motor imagery (MI) are now well-established in different populations of persons suffering from central nervous system impairments. However, research on similar efficacy of MI interventions after amputation remains scarce, and experimental studies were primarily designed to explore the effects of MI after upper-limb amputations. OBJECTIVES: The present comparative study therefore aimed to assess the effects of MI on locomotion recovery following unilateral lower-limb amputation. METHODS: Nineteen participants were assigned either to a MI group (n = 9) or a control group (n = 10). In addition to the course of physical therapy, they respectively performed 10 min per day of locomotor MI training or neutral cognitive exercises, five days per week. Participants' locomotion functions were assessed through two functional tasks: 10 m walking and the Timed Up and Go Test. Force of the amputated limb and functional level score reflecting the required assistance for walking were also measured. Evaluations were scheduled at the arrival at the rehabilitation center (right after amputation), after prosthesis fitting (three weeks later), and at the end of the rehabilitation program. A retention test was also programed after 6 weeks. RESULTS: While there was no additional effect of MI on pain management, data revealed an early positive impact of MI for the 10 m walking task during the pre-prosthetic phase, and greater performance during the Timed Up and Go Test during the prosthetic phase. Also, a lower proportion of participants still needed a walking aid after MI training. Finally, the force of the amputated limb was greater at the end of rehabilitation for the MI group. CONCLUSION: Taken together, these data support the integration of MI within the course of physical therapy in persons suffering from lower-limb amputations.

3D printed orthopedic prostheses for domestic and wild birds-case reports.

Regardless of the species, birds are exposed to injuries that lead to amputation of part of the body structure and often euthanasia. Based on the need for new technologies that improve the quality of life of birds with locomotor problems, the present case reports aimed to describe the development of custom-made three-dimensional (3D) prostheses for domestic and wild birds that suffered amputation or malformation of the hind limb. Using the measurements of the bird, a digital model was created for 3D printing using fused deposition modeling technology (FDM) by the Brazilian company 3D Medicine. In this study we report the use of 3D prosthesis for the rehabilitation of three birds with locomotor disorders in Brazil, the animals adapted to the custom-made prosthesis with an improvement in quality of life, better distribution of body weight, locomotion, and landing. This study describes the development of 3D prostheses for birds in Brazil, the first report of this technology for these species, and the pioneering development of socket prostheses for small birds. 3D prostheses offer a high-efficiency solution to improve the quality of life of animals with amputations and malformations of the hind limbs. In addition, 3D technology provides valuable tools for veterinary medicine, developing custom-made models for the most different anatomical demands of animal patients.

Exploration and demonstration of explainable machine learning models in prosthetic rehabilitation-based gait analysis.

Quantitative gait analysis is important for understanding the non-typical walking patterns associated with mobility impairments. Conventional linear statistical methods and machine learning (ML) models are commonly used to assess gait performance and related changes in the gait parameters. Nonetheless, explainable machine learning provides an alternative technique for distinguishing the significant and influential gait changes stemming from a given intervention. The goal of this work was to demonstrate the use of explainable ML models in gait analysis for prosthetic rehabilitation in both population- and sample-based interpretability analyses. Models were developed to classify amputee gait with two types of prosthetic knee joints. Sagittal plane gait patterns of 21 individuals with unilateral transfemoral amputations were video-recorded and 19 spatiotemporal and kinematic gait parameters were extracted and included in the models. Four ML models-logistic regression, support vector machine, random forest, and LightGBM-were assessed and tested for accuracy and precision. The Shapley Additive exPlanations (SHAP) framework was applied to examine global and local interpretability. Random Forest yielded the highest classification accuracy (98.3%). The SHAP framework quantified the level of influence of each gait parameter in the models where knee flexion-related parameters were found the most influential factors in yielding the outcomes of the models. The sample-based explainable ML provided additional insights over the population-based analyses, including an understanding of the effect of the knee type on the walking style of a specific sample, and whether or not it agreed with global interpretations. It was concluded that explainable ML models can be powerful tools for the assessment of gait-related clinical interventions, revealing important parameters that may be overlooked using conventional statistical methods.