University of Pittsburgh engineers have developed the AR-Based EEG-Guided Neglect Detection System (AREEN), an innovative tool designed to enhance the detection, assessment, and rehabilitation of patients with unilateral spatial neglect (SN) caused by stroke. This novel technology leverages augmented reality (AR) and electroencephalography (EEG) to provide real-time, accurate detection and quantification of neglect, offering significant advancements over traditional methods.
Description
The AREEN system integrates AR and EEG technologies to continuously map and assess patients' neglected visual fields, providing immediate feedback and facilitating effective rehabilitation. By utilizing the Microsoft HoloLens and EEG, AREEN offers a unique composite of real-world and digital environments to improve patient outcomes.
Applications
- Unilateral Spatial Neglect Detection
- Rehabilitation
- Neuropsychological Assessment
Advantages
The use of AR allows for a seamless integration of real-world and digital environments, which is crucial for assessing and rehabilitating spatial neglect in a manner that is directly translatable to daily living activities. Unlike VR-based systems, which can be disorienting or cause motion sickness, AR provides a balanced immersion that better simulates real-world conditions. The inclusion of EEG enhances the system’s diagnostic capabilities by providing a continuous stream of neuroimaging data that can detect and quantify neglect with greater precision and reduced human error. This combination of technologies not only improves the accuracy of SN assessment but also enables the customization of rehabilitation exercises based on real-time feedback, potentially accelerating recovery.
Invention Readiness
AREEN is currently at the prototype stage, with extensive testing validating its effectiveness. The system uses Microsoft HoloLens and Unity Plus to create an AR environment where patients interact with dynamically appearing visual targets. EEG data is simultaneously collected to monitor brain activity, which is then analyzed using a custom-built interface on MATLAB. The prototype has shown promising results in providing both binary diagnoses and graded assessments of neglect severity.
IP Status
https://patents.google.com/patent/WO2023192470A1
