scholarly journals Determining Grasp Selection from Arm Trajectories via Deep Learning to Enable Functional Hand Movement in Tetraplegia

2020 ◽  
Author(s):  
Nikunj Bhagat ◽  
Kevin King ◽  
Richard Ramdeo ◽  
Adam Stein ◽  
Chad Bouton
Keyword(s):  
Spinal Cord ◽  
Spinal Cord Injury ◽  
Deep Learning ◽  
Inertial Sensor ◽  
Hand Movement ◽  
Three Dimensional ◽  
High Accuracy ◽  
Reaching Movements ◽  
Neuromuscular Stimulation ◽  
Cord Injury

Abstract Background: Cervical spinal cord injury severely affects grasping ability of its survivors. Fortunately, many individuals with tetraplegia retain residual arm movements that allow them to reach for objects. We propose a wearable technology that utilizes arm movement trajectory information and deep learning methods to determine grasp selection. Furthermore, we combined this approach with neuromuscular stimulation to determine if self-driven functional hand movement could be enabled in spinal cord injury participants.Methods: Two cervical SCI participants performed arbitrary and natural reaching movements toward target objects in three-dimensional space, which were recorded using an inertial sensor worn on their wrist. Time series classifiers were trained to recognize the trajectories using either a Dynamic Time Warping (DTW) algorithm or a Long Short-Term Memory (LSTM) recurrent neural network. As an initial proof-of-concept, we demonstrate real-time classification of the arbitrary movements using DTW only (due to its implementation simplicity), which when used in combination with a high density neuromuscular stimulation sleeve with textile electrodes, enabled participants to perform functional grasping.Results: Participants were able to consistently perform arbitrary two-dimensional and three-dimensional arm movements which could be classified with high accuracy. Furthermore, it was found that natural reaching trajectories for two different target objects (requiring two different grasp types) were distinct and also discriminable with high accuracy. In offline comparisons, LSTM (mean accuracies 99%) performed significantly better than DTW (mean accuracies 86% and 83%) for both arbitrary and natural reaching movements, respectively. Type I and II errors occurred more frequently for DTW (up to 60% and 15%, respectively), whereas it stayed under 5% for LSTM. Also, DTW achieved online accuracy of 79%. Conclusions: We demonstrate the feasibility of utilizing arm trajectory information to determine grasp selection using a wearable inertial sensor along with DTW and deep learning methods. Importantly, this technology can be successfully used to control neuromuscular stimulation and restore functional independence to individuals living with paralysis.Trial registration: NCT, NCT03385005. Registered September 26, 2017, https://clinicaltrials.gov/ct2/show/NCT03385005

scholarly journals Inferring Grasp Intentions from Arm Trajectories via Deep Learning to Enable Functional Movement in Quadriplegia Revised Title Determining Grasp Selection from Arm Trajectories via Deep Learning to Enable Functional Hand Movement in Tetraplegia

2020 ◽  
Author(s):  
Nikunj Bhagat ◽  
Kevin King ◽  
Richard Ramdeo ◽  
Adam Stein ◽  
Chad Bouton
Keyword(s):  
Spinal Cord ◽  
Spinal Cord Injury ◽  
Deep Learning ◽  
Inertial Sensor ◽  
Hand Movement ◽  
Three Dimensional ◽  
High Accuracy ◽  
Reaching Movements ◽  
Neuromuscular Stimulation ◽  
Cord Injury

Abstract Background: Cervical spinal cord injury severely affects grasping ability of its survivors. Fortunately, many individuals with tetraplegia retain residual arm movements that allow them to reach for objects. We propose a wearable technology that utilizes arm movement trajectory information and deep learning methods to determine grasp selection. Furthermore, we combined this approach with neuromuscular stimulation to determine if self-driven functional hand movement could be enabled in spinal cord injury participants.Methods: Two cervical SCI participants performed arbitrary and natural reaching movements toward target objects in three-dimensional space, which were recorded using an inertial sensor worn on their wrist. Time series classifiers were trained to recognize the trajectories using either a Dynamic Time Warping (DTW) algorithm or a Long Short-Term Memory (LSTM) recurrent neural network. As an initial proof-of-concept, we demonstrate real-time classification of the arbitrary movements using DTW only (due to its implementation simplicity), which when used in combination with a high density neuromuscular stimulation sleeve with textile electrodes, enabled participants to perform functional grasping.Results: Participants were able to consistently perform arbitrary two-dimensional and three-dimensional arm movements which could be classified with high accuracy. Furthermore, it was found that natural reaching trajectories for two different target objects (requiring two different grasp types) were distinct and also discriminable with high accuracy. In offline comparisons, LSTM (mean accuracies 99%) performed significantly better than DTW (mean accuracies 86% and 83%) for both arbitrary and natural reaching movements, respectively. Type I and II errors occurred more frequently for DTW (up to 60% and 15%, respectively), whereas it stayed under 5% for LSTM. Also, DTW achieved online accuracy of 79%. Conclusions: We demonstrate the feasibility of utilizing arm trajectory information to determine grasp selection using a wearable inertial sensor along with DTW and deep learning methods. Importantly, this technology can be successfully used to control neuromuscular stimulation and restore functional independence to individuals living with paralysis.

scholarly journals Cerebellar contribution to sensorimotor adaptation deficits in humans with spinal cord injury

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Yuming Lei ◽  
Monica A. Perez
Keyword(s):  
Spinal Cord ◽  
Spinal Cord Injury ◽  
Hand Movement ◽  
Cerebellar Atrophy ◽  
Reaching Movements ◽  
Sensorimotor Adaptation ◽  
Post Injury ◽  
Control Subjects ◽  
Learning Rates ◽  
Cord Injury

AbstractHumans with spinal cord injury (SCI) show deficits in associating motor commands and sensory feedback. Do these deficits affect their ability to adapt movements to new demands? To address this question, we used a robotic exoskeleton to examine learning of a sensorimotor adaptation task during reaching movements by distorting the relationship between hand movement and visual feedback in 22 individuals with chronic incomplete cervical SCI and 22 age-matched control subjects. We found that SCI individuals showed a reduced ability to learn from movement errors compared with control subjects. Sensorimotor areas in anterior and posterior cerebellar lobules contribute to learning of movement errors in intact humans. Structural brain imaging showed that sensorimotor areas in the cerebellum, including lobules I–VI, were reduced in size in SCI compared with control subjects and cerebellar atrophy increased with increasing time post injury. Notably, the degree of spared tissue in the cerebellum was positively correlated with learning rates, indicating participants with lesser atrophy showed higher learning rates. These results suggest that the reduced ability to learn from movement errors during reaching movements in humans with SCI involves abnormalities in the spinocerebellar structures. We argue that this information might help in the rehabilitation of people with SCI.

scholarly journals Inferring Grasp Intentions from Arm Trajectories via Deep Learning to Enable Functional Movement in Quadriplegia

2020 ◽  
Author(s):  
Nikunj Bhagat ◽  
Kevin King ◽  
Richard Ramdeo ◽  
Chad Bouton
Keyword(s):  
Deep Learning ◽  
Short Term Memory ◽  
Inertial Sensor ◽  
Cervical Spinal Cord ◽  
Wearable Sensors ◽  
Functional Independence ◽  
Reaching Movements ◽  
Type I ◽  
Neuromuscular Stimulation ◽  
Cord Injury

Abstract Background Cervical spinal cord injury severely affects grasping ability of its survivors. Fortunately, many individuals with quadriplegia retain residual arm movements that allow them to reach for objects. We propose a wearable technology that utilizes pattern recognition and deep learning methods to automatically classify arm trajectories and infer grasping intentions. Further, this technology can enable individuals with SCI to grasp objects without assistance via neuromuscular stimulation. Methods Two cervical SCI participants performed various reaching movements and smooth trajectories in space, which were recorded using an inertial sensor worn on their wrist. Time series classifiers were trained to recognize the trajectories using either a Dynamic Time Warping (DTW) algorithm or a Long Short-Term Memory (LSTM) recurrent neural network. Successful trajectory prediction in real-time was demonstrated using DTW, which when used in combination with a high density neuromuscular stimulation sleeve with textile electrodes, enabled participants to perform functional grasps. Results In offline comparisons, LSTM (mean accuracies 98% and 99%) performed significantly better than DTW (mean accuracies 94% and 83%) for both 2D and 3D reaching movements, respectively. Type I and II errors occurred more frequently for DTW (up to 38% and 15%, respectively), whereas it stayed under 3% for LSTM. Also, DTW achieved online accuracy of 79 ± 5%. Conclusions We demonstrate the feasibility of inferring grasping intention from reaching trajectories using wearable sensors. Importantly, this technology can be successfully used to control neuromuscular stimulators and restore functional independence to individuals living with paralysis. Trial registration: NCT, NCT03385005. Registered September 26, 2017, https://clinicaltrials.gov/ct2/show/NCT03385005

Transplantation of collagen sponge-based three-dimensional neural stem cells cultured in the RCCS system facilitate locomotor functional recovery in spinal cord injury animals

2022 ◽  
Author(s):  
Jianwu Dai ◽  
Yunlong Zou ◽  
Yanyun Yin ◽  
Zhifeng Xiao ◽  
Yannan Zhao ◽  
...  
Keyword(s):  
Spinal Cord ◽  
Tissue Engineering ◽  
Stem Cells ◽  
Spinal Cord Injury ◽  
Neural Stem Cells ◽  
Three Dimensional ◽  
Collagen Sponge ◽  
Cellular Functions ◽  
Cord Injury ◽  
Cells Cultured

Numerous studies have indicated that microgravity induces various changes in the cellular functions of neural stem cells (NSCs), and the use of microgravity to culture tissue engineering seed cells for...

Sign in / Sign up

Export Citation Format

Share Document