Designing a pervasive eye movement-based system for ALS and paralyzed patients

Detecting human eyes movement and using it for communication is very common for people with amyotrophic lateral sclerosis (ALS) and other locked-in and paralysis diseases. The majority of existing systems are very expensive and nearly all of them use special devices and cameras, mostly infrared based, that are not commonly available. For Persian patients there is no special user interface at all. The system introduced in this paper which is called EyeType uses just a normal webcam and a calibration-free algorithm to identify eye gestures and by using a specific user interface enables the patients to type what they want. To make it easier, a pervasive algorithm is applied to score the words that are mostly used and after collecting the required information, begins to suggest most possible words to the user as he/she types the initial letters.