NBNikunj Arunkumar Bhagat

Nikunj Arunkumar Bhagat

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Nikunj Arunkumar Bhagat has a total of 70 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Assessing the Utility of Fronto-Parietal and Cingulo-Opercular Networks in Predicting the Trial Success of Brain-Machine Interfaces for Upper Extremity Stroke Rehabilitation

    by Gokul Krishna Raja Padmaja, Nikunj Arunkumar Bhagat, Pragathi Priyadharsini Balasubramani

    Brain-machine interfaces (BMIs) have the potential to improve stroke rehabilitation by actively facilitating sensory-cognitive-motor connections to restore movement. However, individuals with cognitive impairments are often excluded from BMI-based neurorehabilitation due to concerns about impaired cognition, specifically reduced attention and executive control. We propose leveraging the trial-wise dynamics of large-scale cognitive control networks-specifically, the frontoparietal (FPN) and cingulo-opercular (CON) networks-to build neural markers of cognitive control. Using existing BMI datasets, we demonstrate that trial-wise activity within these networks predicts motor task performance, suggesting that cognitive control signals in these networks could serve as adaptive modulations for BMI-based rehabilitation. Our system is able to predict unsuccessful BMI trials at the population level about 84.2% of the time on average, with an overall mean accuracy of 72.2% in a 3-fold cross-validation. Additionally, in a leave-one-subject-out validation, our system achieved 71% specificity on average, with an overall mean accuracy of 68.3%. Notably, model performance varies across subjects, with some individuals showing up to 92% specificity and 100% sensitivity. Unlike previous studies that primarily focus on resting-state data, our findings point toward the untapped potential of incorporating cognitive network state monitoring into BMI systems to optimize online performance through trials. Specifically, we suggest that our pre-trained models can be fine-tuned with subject-specific information to design more targeted rehabilitation programs that enhance motor performance by identifying precise attention and learning tasks to improve the successful response of the network model in patients with significant cognitive impairment.


  • A State-Space Control Approach for Tracking Isometric Grip Force During BMI Enabled Neuromuscular Stimulation

    by Nikunj A. Bhagat, Gerard E. Francisco, Jose L. Contreras-Vidal

    Sixty percent of elderly hand movements involve grasping, which is unarguably why grasp restoration is a major component of upper-limb rehabilitation therapy. Neuromuscular electrical stimulation is effective in assisting grasping, but challenges around patient engagement and control, as well as poor movement regulation due to fatigue and muscle nonlinearity continue to hinder its adoption for clinical applications. In this study, we integrate an electroencephalography-based brain–machine interface (BMI) with closed-loop neuromuscular stimulation to restore grasping and evaluate its performance using an isometric force tracking task. After three sessions, it was concluded that the normalized tracking error during closed-loop stimulation using a state-space feedback controller (25 ± 15%), was significantly smaller than conventional open-loop stimulation (31 ± 24%), ( F (748.03, 1) = 23.22, p < 0.001). Also, the impaired study participants were able to achieve a BMI classification accuracy of 65 ± 10% while able-bodied participants achieved 57 ± 18% accuracy, which suggests the proposed closed-loop system is more capable of engaging patients for rehabilitation. These findings demonstrate the multisession performance of model-based feedback-controlled stimulation, without requiring frequent reconfiguration.


  • Open Access

    A State-Space Control Approach for Tracking Isometric Grip Force during BMI enabled Neuromuscular Stimulation

    by Nikunj Bhagat, Gerard Francisco, Jose Contreras-Vidal

    This paper develops a feedback-controlled neurmuscular stimulator focussed on hand rehabilitation, particularly grasping, which was tested in a multi-session study with paralyzed patients. We also evaluate the accuracy of controlling neuromuscular stimulation devices using EEG-based brain-machine interfaces. “This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible.”

  • Open Access

    Targeted transcutaneous cervical spinal cord stimulation promotes upper limb recovery in spinal cord and peripheral nerve injury

    by Santosh Chandrasekaran, Nikunj A Bhagat, Richard Ramdeo, Sadegh Ebrahimi, Pawan D Sharma, Adam Stein, Doug G Griffin, Susan J Harkema, Chad E Bouton

    Long-term recovery of limb function is a significant unmet need in people with paralysis. Targeted neuromodulation through epidural stimulation of the spinal cord, when paired with intense physical therapy, has shown promising results towards restoring volitional control of the upper limb in people with tetraplegia. Non-invasive neuromodulation of the cervical spinal cord using transcutaneous spinal cord stimulation (tSCS) has shown similar improvements in upper-limb motor control rehabilitation. However, the potential rehabilitative effects of activating specific cervical spinal segments in a targeted fashion using tSCS has largely remained unexplored. We show in two individuals with SCI that tSCS of the cervical spinal cord resulted in up to 1136% increase in exerted force, with minimal physical therapy. We also show, for the first time, the effectiveness of targeted tSCS in restoring strength (407% increase) and dexterity to the digits in an individual with paralysis of the hand due to a peripheral nerve injury. Furthermore, we believe this is the first study to document a 3-point improvement in sensation in people with SCI or peripheral injury after receiving tSCS. Lastly, participant gains persisted after a one-month period void of stimulation, suggesting tSCS may lead to lasting benefits. Non-invasive targeted spinal cord stimulation shows tremendous promise as a safe and effective therapeutic approach with broad applications.


  • Open Access

    A Roadmap Towards Standards for Neurally Controlled End Effectors

    by Andrew Y. Paek, Justin A. Brantley, Akshay Sujatha Ravindran, Kevin Nathan, Yongtian He, David Eguren, Jesus G. Cruz-Garza, Sho Nakagome, Dilranjan S. Wickramasuriya, Jiajun Chang, et al.

    The control and manipulation of various types of end effectors such as powered exoskeletons, prostheses, and ‘neural’ cursors by brain-machine interface (BMI) systems has been the target of many research projects. A seamless “plug and play” interface between any BMI and end effector is desired, wherein similar user's intent cause similar end effectors to behave identically. This report is based on the outcomes of an IEEE Standards Association Industry Connections working group on End Effectors for Brain-Machine Interfacing that convened to identify and address gaps in the existing standards for BMI-based solutions with a focus on the end-effector component. A roadmap towards standardization of end effectors for BMI systems is discussed by identifying current device standards that are applicable for end effectors. While current standards address basic electrical and mechanical safety, and to some extent, performance requirements, several gaps exist pertaining to unified terminologies, data communication protocols, patient safety and risk mitigation.


  • Open Access

    Evoking highly focal percepts in the fingertips through targeted stimulation of sulcal regions of the brain for sensory restoration

    by Santosh Chandrasekaran, Stephan Bickel, Jose L Herrero, Joo-Won Kim, Noah Markowitz, Elizabeth Espinal, Nikunj A Bhagat, Richard Ramdeo, Junqian Xu, Matthew F Glasser, et al.

    Paralysis and neuropathy, affecting millions of people worldwide, can be accompanied by a significant loss of somatosensation. With tactile sensation being central to achieving dexterous movement, brain-computer interface (BCI) researchers have explored the use of intracortical electrical stimulation to restore sensation to the hand. However, current approaches have been restricted to stimulating the gyral areas of the brain while functional imaging suggests that the representation of fingertips lie predominantly in the sulcal regions. Here we show, for the first time, highly focal percepts can be evoked in the fingertips of the hand through electrical stimulation of the sulcal areas of the brain. To this end, we mapped and compared sensations elicited in the hand by stimulating both gyral and sulcal areas of the human primary somatosensory cortex (S1). Two participants with intractable epilepsy were implanted with stereoelectroencephalography (SEEG) and high-density electrocorticography (HD-ECoG) electrodes in S1 guided by high-resolution functional imaging. Using myelin content and cortical thickness maps developed by the Human Connectome Project, we elucidated the specific sub-regions of S1 where focal percepts were evoked. Within-participant comparisons showed that sulcal stimulation using SEEG electrodes evoked percepts that are significantly more focal, with 80% less area of spread (p=0.02) and localized to the fingertips more often than in gyral stimulation via HD-ECoG electrodes. Finally, sulcal locations exhibiting repeated modulation patterns of high-frequency neural activity during mechanical tactile stimulation of the hand showed the same somatotopic correspondence as sulcal stimulation. These findings show that minimally-invasive sulcal stimulation could lead to a clinically viable approach to restoring sensation in those living with sensory impairment. Significance Intracortical or cortical surface stimulation of the primary somatosensory cortex (S1) offers the promise of restoring somatotopically-relevant sensation in people with sensory impairment. However, evoking percepts in the fingertips has been challenging as their representation has been shown to be predominantly located within sulcal regions of S1 – inaccessible by these stimulation approaches. We evoked highly focal percepts in the fingertips of the hand by stimulating the sulcal regions of S1 in people with intractable epilepsy using stereoelectroencephalography (SEEG) depth electrodes. Sensory percepts in the fingertips were more focal and more frequently evoked by SEEG electrodes than by high-density electrocorticography (HD-ECoG) grids evidenced by within-participant comparisons. Our results suggest that fingertip representations are more readily targeted within the sulcal regions. SEEG electrodes potentially offer a clinically viable approach to access the sulcal regions for sensory neuroprostheses that can aid dexterous motor control.


  • Open Access

    Determining grasp selection from arm trajectories via deep learning to enable functional hand movement in tetraplegia

    by Nikunj Bhagat, Kevin King, Richard Ramdeo, Adam Stein, Chad Bouton

    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


  • Open Access

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

    by Nikunj Bhagat, Kevin King, Richard Ramdeo, Adam Stein, Chad Bouton

    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

  • Open Access

    A clinical trial to study changes in neural activity and motor recovery following brain-machine interface enabled robot-assisted stroke rehabilitation

    by Nikunj Arunkumar Bhagat, Nuray Yozbatiran, Jennifer L. Sullivan, Ruta Paranjape, Colin Losey, Zachary Hernandez, Zafer Keser, Robert Grossman, Gerard Francisco, Marcia K. O’malley, et al.

    Background: Brain-machine interfaces (BMI) based on scalp electroencephalography (EEG) have the potential to promote cortical plasticity following stroke, which has been shown to improve motor recovery outcomes. However, clinical efficacy of BMI-enabled robotic rehabilitation in chronic stroke population is confounded by the spectrum of motor impairments caused by stroke. Objective: To evaluate the efficacy of neurorehabilitation therapy on upper-limb motor recovery, by quantifying changes in clinical, BMI-based, and kinematics-based metrics. Further, to identify neural correlates or biomarkers that can predict the extent of motor recovery. Methods: Chronic stroke survivors (n = 10, age 55 ± 9.2y, chronicity 3.1 ± 2.8y) were recruited to participate in a 4-6 weeks long clinical study. Participants completed 12 therapy sessions that involved a BMI enabled powered exoskeleton (MAHI Exo-II) for training, which targeted elbow flexion and extension. Clinical assessments including Fugl-Meyer Upper Extremity (FMA-UE) and Action Research Arm Test (ARAT) were measured up to 2-months after therapy. BMI performance, kinematic performance, and change in movement related cortical potentials (MRCP) were also determined. Results: On average, 132 ± 22 repetitions were performed per participant, per session. BMI accuracy across all sessions and subjects was 79 ± 18%, with a small number of false positives (23 ± 20%). FMA-UE and ARAT scores improved significantly over baseline after therapy and were retained at follow-ups (ΔFMA-UE = 3.92 ± 3.73 and ΔARAT = 5.35 ± 4.62, p < 0.05). 80% participants (7 with moderate-mild impairment and 1 with severe-moderate impairment) reached minimal clinically important difference (MCID: FMA-UE > 5.2 or ARAT > 5.7) during the course of the study. Kinematic measures indicate that, on average, participants’ movements became faster and smoother. Quantification of changes in MRCP amplitude showed significant correlation with ARAT scores (ρ = 0.72, p < 0.05) and marginally significant correlation with FMA-UE (ρ = 0.63, p = 0.051), suggesting higher activation of ipsi-lesional hemisphere post-intervention. The study did not have any adverse events. Conclusion: This study presents evidence that BMI enabled robotic rehabilitation can promote motor recovery in individuals with chronic stroke, several years after injury and irrespective of their impairment level, or location of the lesion (cortical/subcortical) at baseline. Further, the extent of motor recovery was correlated with changes in movement related potentials, occurring contralateral to the impaired arm. Support: NIH National Robotics Initiative Grant R01NS081854 and a grant from Mission Connect, a project of TIRR Foundation.


  • Open Access

    Neural activity modulations and motor recovery following brain-exoskeleton interface mediated stroke rehabilitation

    by Nikunj A. Bhagat, Nuray Yozbatiran, Jennifer L. Sullivan, Ruta Paranjape, Colin Losey, Zachary Hernandez, Zafer Keser, Robert Grossman, Gerard E. Francisco, Marcia K. O'MAlley, et al.