Ostadabbas Receives an NSF PFI Grant to Integrate AR Technologies into Stroke Rehabilitation

Sarah Ostadabbas

ECE Associate Professor Sarah Ostadabbas, in collaboration with the University of Pittsburgh and Myomo, Inc., has secured a $550K NSF grant for their project titled “PFI-RP: Augmented Reality and Electroencephalography for Detecting, Assessing, and Rehabilitating Visual Unilateral Neglect in Stroke Patients.” This project aims to create a comprehensive tool for detecting, assessing, and rehabilitating neglect in stroke patients. It will use augmented reality (AR) and electroencephalography (EEG) to automatically detect neglect and stimulate the affected side of the body and environment. This stimulation will target the neural networks responsible for spatial attention and perceptual processing, fostering implicit learning through repeated stimulation during daily activities. This approach aims to support the generalization of learning and enhance the independence of stroke patients.

Abstract Source: NSF

The broader impact/commercial potential of this Partnerships for Innovation Research Partnership (PFI-RP) project is in the development of a visual unilateral neglect detection, assessment, and rehabilitation tool for stroke patients. According to the Center for Disease Control and Prevention, stroke is a leading cause of serious long-term disability, including the cost of healthcare services, medicines and missed days of work. Stroke related costs in the United States are around $50 billion every year. Unilateral spatial neglect (neglect) is a disabling neurocognitive impairment for stroke patients occurring in 28.60% of the stroke population. Moreover, neglect is a strong predictor of disability, and stroke patients with neglect have longer hospital stays and receive more direct treatment time from physical and occupational therapists than individuals without neglect, culminating in significantly higher healthcare costs. However, there is currently no commercial product targeting neglect detection, assessment and rehabilitation. Current clinical gold standard neglect detection methods are affected by compensatory strategies and are not sensitive to changes in the severity of neglect; and rehabilitation techniques are not effective in reducing disability associated with neglect. This project, through a multidisciplinary team, will address these issues while developing the proposed neglect detection, assessment, and rehabilitation technology.

The proposed project will develop a neglect detection, assessment and rehabilitation tool that incorporates stimulation to draw attention to the affected side of the body and environment based on the automatic detection of neglect through neurophysiology measured through electroencephalography, thereby stimulating neural networks responsible for spatial attention and perceptual processing for stroke patients. Stimulation during everyday activities with high levels of repetition is needed to promote implicit learning that supports generalization of learning and independence. Current detection and rehabilitation methods lack the necessary precision, salience, and repetition to promote this type of learning. Due to limitations associated with these methods, there is no gold assessment for neglect or standard of care for rehabilitation of neglect in routine clinical practice. The proposed project offers a potential solution to these problems, incorporating best practices from rehabilitation intervention research and state-of-the-art technologies that will be cost-effective to be implemented in rehabilitation clinics. The proposed technology incorporates objective and reliable quantification of neglect severity and tailored sensory stimulation that is: (i) technology-driven (accurate, reliable, repetitive, cost-effective); (ii) multimodal: visual, auditory, haptic stimulation (optimizes perceptual processing); (iii) individualized to the available field of vision; and (iv) simultaneous with “real-world” activities of daily living .

Related Faculty: Sarah Ostadabbas

Related Departments:Electrical & Computer Engineering