An Intelligent Human-Machine Interface Based on Eye Tracking for Communication of Patients with Locked in Syndrome
M M BHATHRISHA · INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT · 2025
Paralyzed people lack the ability to control muscle function in one or more muscle groups. The condition can be caused by stroke, ALS, multiple sclerosis, and many other diseases. Locked in Syndrome (LIS) is a form of paralysis where patients have lost control of nearly all voluntary muscles. These people are unable to control any part of their body, besides eye movement and blinking. Due to their condition, these people are unable to talk, text, and communicate in general. Even though people that have LIS are cognitively aware, their thoughts and ideas are locked inside of them. These people depend on eye blinks to communicate. They rely on nurses and caretakers to interpret and decode their blinking. Whenever LIS patients do not have a person to read their eye blinks available, they have no means of self-expression. Our project Blink to Text offers a form of independence to paralyzed people. The software platform converts eye blinks to text. Every feature of the software can be controlled by eye movement. Thus, the software can be independently operated by paralyzed people. The software can be run on any low-end computer. The software uses computer vision and Haar cascades to detect eye blinking and convert the motion into text. The program uses language modelling to predict the next words that the user might blink. INDEX TERMS: Eye-tracking, Deep learning, human machine interface.