A Blink-based Communication (BLINC) System for Patients in Bed

Period of Performance: 09/01/2015 - 08/31/2016

$698K

Phase 2 SBIR

Recipient Firm

Barron Assoc., Inc.
CHARLOTTESVILLE, VA 22901
Principal Investigator
Principal Investigator

Abstract

DESCRIPTION (provided by applicant): Introduction: The Blink-based Communication (BLINC) System will allow any patient who retains oculomotor control to reliably signal for assistance and communicate basic requests to a caregiver. It uses infrared (IR) cameras and illuminators to continuously monitor a patient in bed, and to detect blink signals even in a dark room. It is specifically designed so that all components, including the cameras and illuminators, can be placed at a significant standoff distance, making it easy to use and minimally intrusive. The system is initially targeted to help ALS patients, who commonly lose virtually all voluntary muscle control but frequently do retain oculomotor control. The BLINC System will provide access to basic and essential communication capabilities that can be extremely difficult or impossible to provide with existing technology, with an ultimate goal of improving quality of life for severely impaired patients. Problems Addressed: The BLINC System will not require any components to be placed in close proximity to the patient, or to be positioned or configured each time a patient is placed in bed or repositioned. The system addresses a significant capability gap in current high-end Augmentative and Alternative Communication (AAC) devices, which effectively leverage oculomotor capabilities for communication, but require the device to be placed directly in front of the patient's face and in close proximity, which is not practical in al settings. Specialized buttons and switches can provide a simple call bell, but these require a degree of muscle control and must be carefully placed in close proximity each time a patient is moved. Phase I Results: The Phase I e ort successfully demonstrated the feasibility of the approach through human participant tests involving healthy subjects. A prototype BLINC System was implemented that detected blink signals (a sequence of four blinks of approximately two seconds each) in real-time and activated an audio signal. The tests showed at the 95% confidence level that the system accuracy exceeded the 95% target for Phase I. A specificity of over 99% also demonstrated that intentional blink signals could be reliably discriminated from normal eyelid movements including spontaneous blinks. Phase II Summary: The proposed Phase II e ort will build on the successful Phase I prototype system, expanding the set of communication capabilities, and enhancing the robustness of the system. Enhancements will ensure robustness to a wider range of user poses, user positions, and lighting conditions. Phase II human participant tests will involve ALS patients in two stages, first assessing system accuracy in a very controlled clinic setting, and second assessing usability and false alarm rates in multi-day trials in patient's homes. An important aspect of Phase II will be involving caregiver so that usability can be understood from both the patient and caregiver perspective, a key factor in overall system acceptance.