Research Projects

Rationale:Human upper limb movements show extraordinary diversity and dexterity, which is a result of the rich coordination of muscle activation patterns. Stroke is a devastating neurological injury that disrupts these controls. A majority of stroke survivors routinely suffer from upper extremity impairments due to symptoms such as paresis, spasticity, hypertonia, and impeding arm joint (elbow/shoulder) movements. Disuse atrophy of the muscle results in a further reduction in workspace volume and ease of manipulation during voluntary arm movement, severely affecting the quality of life of stroke survivors. Thus, appropriate intervention is absolutely necessary in a routine manner. Assisted arm movement therapy (AST) is widely used for arm rehabilitation. The planning of appropriate interventions for AST depends upon the specific needs of the patient and on the prevailing deficits. Existing manual clinical assessments of voluntary arm movements are valuable but lack the necessary precision. Furthermore, visual inspection of limb movements is also prone to large estimation errors. Accordingly, accurate assessments of voluntary movement deficits should play an important role in planning and delivering rehabilitation interventions. Rehabilitation of the paretic arm is a resource-intensive dose-dependent factor that requires continuous engagement of the therapist. The global and especially the Indian scenario of availability of a qualified therapist or rehabilitation centers is alarmingly low. Thus creating additional challenges to our citizens' right to health. Objective: To address this gap, in this PM ECR grant, I propose an easy-to-use noninvasive wireless wearable sensor based system for precise measurement of deficits in voluntary reaching movements along with muscle activity of the upper limb that will potentially help to quantify the outcomes of clinical intervention. A system will be peripheral to an Artificial Neural Network-based automatic assessment framework to assess the severity of the deficit and recovery. I hypothesize the automatic prediction will be aligned with the clinical assessments Experiment plan:The wireless system will be used to record the arm movements of hemiparetic stroke survivors. Instructed arm movements will be supervised through a virtual reality interface. Healthy individuals will be recorded for benchmarking, followed by a validation with stroke survivors. The measurements will then automatically estimate the severity of the impairment and the recovery through the ANN-based framework. Significance:This smart, noninvasive wireless sensor-based virtual reality-based framework will potentially precisely measure and estimate voluntary arm movement disorders and muscle activation deficits in stroke survivors. It will also provide automatic prediction of the severity of the impairment. Thus, it will facilitate frequent repetition of the intervention even with future possibilities of telerehabilitation.