Periodic Reporting for period 2 - SimBionics (Neuromechanical Simulation and Sensory Feedback for the Control of Bionic Legs)
Berichtszeitraum: 2021-09-01 bis 2023-08-31
Robotic lower limb prostheses, or bionic legs, have underwent tremendous mechatronic advances in the past decade. However, current prosthesis control interfaces are still sub-optimal leading to inefficient walking. To address this limit, there is need for highly trained professionals that work at the intersection of academic research, clinical assessment and market translation. The main goal of SimBionics is to prepare highly trained professionals that are able to work at the intersection of academic research, clinical assessment and market translation. For this purpose, the consortium has prepared a training platform that provides theoretical as well as practical experience in both the academic and the industrial environment. There is a tight coordinating between the academic and the industrial partners to guarantee that all activities done by the ESRs will contribute to their specific career development plans and accomplish the technical objectives of the project. The scientific objective of the project is to improve the control interfaces for mechatronic prosthetic devices by (1) including balance in the control algorithms for lower limb prosthesis, (2) by designing and employing a volitional control that accurately detects user intentions with a reduced set of sensors (in the absence of electromyographic signals), (3) by providing non-invasive tactile feedback and (4) to create a comprehensive assessment methodology to evaluate the progress and impact of the proposed techniques. SimBionics enhanced the ESRs’ career perspectives and established a reference for technology development training program, by extending the body of knowledge in the field and providing innovative solutions for lower limb amputees.
During the project the ESRs implement a neuro-musculoskeletal (NMS) model of the missing limb, which provides a representation of the equivalent missing muscle and bone structures that are now replaced by prosthetic devices. The model was used as a baseline to implement control strategies for balance and walking using a powered ankle prosthetic foot as well as providing feedback back to the amputee. Importantly, the project also worked on exploring novel ways to assess biomechanical performance as well as cognitive effort of amputees when using a prosthetic device (Fig. 1). All these elements were tested with healthy subjects as well as amputees, showing the feasibility of the concept.
Specifically, a model of muscle synergies capable of generating synthetic EMGs was developed. These synthetic EMGs were used as an input to the NMS model to generate torque output for the prosthetic device as a feedforward control strategy. In addition, an EMG-driven musculoskeletal control was implemented, to allow amputees to volitionally control a prosthetic device directly using their remaining muscles. Furthermore, a model of spinal reflexes was implemented. This model is used to estimate muscle activations by modelling the feedback control strategy that the central nervous system employs to modulate muscle activation in response to environmental conditions and stimuli. This model is used as an input to the NMS model, which generates the torques that the prosthetic device should apply to recover from perturbations. The results were presented and demonstrated in multiple international conferences.
A sensory feedback platform was implemented using vibrotactile and electrotactile actuators. The platform is being used to close the loop between the prosthetic device and the amputee. The feedback platform can convey the information obtained from the sensors of a prosthetic device as well as from the implemented NMS model, which enables more flexibility to find meaningful feedback parameters to improve balance and walking. The results were published in the Journal of Neuroengineering and Rehabilitation as well as multiple international conferences.
Finally, two different methods to assess the use of mechatronic lower limb prosthetic devices were implemented. An experimental study and data analysis to understand the potential of using the internal sensors of mechatronic devices as sources for gait assessment. If possible, this would allow researchers and clinicians to assess the impact of the prosthetic device on the amputee gait without the need for specialized and rather expensive equipment as well as gathering data in the real environment (out of the lab). The results were published in the Journal of Clinical Biomechanics as well as several international conferences. Furthermore, different tools to assess the cognitive attention when walking with mechatronic devices were explored. Several tests investigating the use of mobile eye trackers to measure cognitive load were completed and the results were presented in several international conferences.
A joint IP on the concept of SimBionics, under the title: Method for controlling an orthopedic device and orthopedic device. The patent is now published with numbers: WO2022017583A1, EP4185245A1 and US2023255802A1.
Specifically, a model of muscle synergies capable of generating synthetic EMGs was developed. These synthetic EMGs were used as an input to the NMS model to generate torque output for the prosthetic device as a feedforward control strategy. In addition, an EMG-driven musculoskeletal control was implemented, to allow amputees to volitionally control a prosthetic device directly using their remaining muscles. Furthermore, a model of spinal reflexes was implemented. This model is used to estimate muscle activations by modelling the feedback control strategy that the central nervous system employs to modulate muscle activation in response to environmental conditions and stimuli. This model is used as an input to the NMS model, which generates the torques that the prosthetic device should apply to recover from perturbations. The results were presented and demonstrated in multiple international conferences.
A sensory feedback platform was implemented using vibrotactile and electrotactile actuators. The platform is being used to close the loop between the prosthetic device and the amputee. The feedback platform can convey the information obtained from the sensors of a prosthetic device as well as from the implemented NMS model, which enables more flexibility to find meaningful feedback parameters to improve balance and walking. The results were published in the Journal of Neuroengineering and Rehabilitation as well as multiple international conferences.
Finally, two different methods to assess the use of mechatronic lower limb prosthetic devices were implemented. An experimental study and data analysis to understand the potential of using the internal sensors of mechatronic devices as sources for gait assessment. If possible, this would allow researchers and clinicians to assess the impact of the prosthetic device on the amputee gait without the need for specialized and rather expensive equipment as well as gathering data in the real environment (out of the lab). The results were published in the Journal of Clinical Biomechanics as well as several international conferences. Furthermore, different tools to assess the cognitive attention when walking with mechatronic devices were explored. Several tests investigating the use of mobile eye trackers to measure cognitive load were completed and the results were presented in several international conferences.
A joint IP on the concept of SimBionics, under the title: Method for controlling an orthopedic device and orthopedic device. The patent is now published with numbers: WO2022017583A1, EP4185245A1 and US2023255802A1.
SimBionics has generated several outcomes in terms of scientific, patents, and technical knowledge in all the domains addressed by the project. These outcomes have pushed substantially the state of the art in prosthetic control by proposing a new way of thinking in the field and promoting discussion on new ways to improve prosthetic control. The overall progress beyond the state of the art is a real time closed-loop controller that integrates a walking controller, a balance controller and a feedback interface into a mechatronic prosthetic device for lower limb amputees. Importantly, the project explored new assessment methodologies to provide quantitative methods and metrics that allows a more comprehensive assessment of the physical and cognitive impact of a lower limb prosthesis on the amputee. These new methods provide a first step towards a more accurate assessment of the functional capabilities and limitations offered by a prosthetic device. This information could be used in the future to help clinicians and reimbursement organizations to better match the amputee with a prosthesis. Importantly, the project promoted discussion on topics related to user needs and what is really needed to improve the quality of living of amputees. Also, promoted the discussion about the difficulties, limitations and opportunities to bring new high-technology to the market in this field.