Eye tracking technology has been continuously researched for application in various fields. In the past, studies have been conducted to interpret eye movements in 3D space in order to solve the problem of not being able to find the centre of rotation of the eye. In this paper, we propose a novel pre-processing method for eye-gaze tracking by monitoring the front of the face with a camera. Our method works regardless of the distance between the eye and the camera. The proposed method includes an analysis technique that simplifies conventional three-dimensional space analysis to two dimensions. The contribution this work presents is a method to simplify gaze direction detection. The errors in our model’s estimations appear to be under 1 pixel. In addition, our approach has an execution time of less than 1 s, enabling an adaptive model that responds to user movements in real time. The proposed method was able to overcome various problems that methods in existing studies still suffer from, including accurately finding the rotational centre of the user’s eye-ball. Moreover, even when a user’s pupil can only be monitored from a distance, our approach still makes it possible to produce accurate estimations.
Background: The biggest problem in an aging society is the development of degenerative brain disease in the elderly. Neurodegenerative brain disease can cause cognitive dysfunction and rapidly increase the prevalence of dementia and Parkinson's disease, posing a huge economic and social burden on the elderly. A computerized cognitive rehabilitation training system has been developed to prevent and train cognitive dysfunction, showing various clinical effects. However, few studies have analyzed components of contents such as memory and concentration training. In this study, the clinical effects and characteristics of the color, number and words elements were analyzed by subdividing the memory and concentration contents into elements, difficulty, and training methods.Methods: Using a total of eight contents developed based on neuropsychology, 24 normal subjects with an average age of 60.58 ± 3.96 years were conducted 3 times a week, and training was received for 30 to 45 minutes per session. To determine the training effect, MMSE-K, an evaluation tool most closely related to cognitive therapy, was used. The number of errors and problem solving time used in the analysis were dataized by measuring the number of incorrect answers selected by the subject and the time spent solving the problem, respectively. Using t-test, the significance of different between before and after training was determined. Correlation between the number of errors and problem-solving time by week was determined using a trend line. All experimental procedures and evaluations were conducted after obtaining IRB approval from Dongguk University Ilsan Hospital (DUIH2020-07-001).Results: The subjects' MMSE-K scores were 27.88 ± 1.70 points before intervention to 28.63 ± 1.69 points after three weeks of intervention. In each subdivided component, color element showed an effect of improving complex difficulty, number element had the most effective training effect, and word element had a predictive effect on cognitive decline. Conclusions: A detailed analysis results of the components used in a computerized cognitive rehabilitation training system will help develop degenerative brain disease contents to be developed later, and is expected to contribute to a prevention-oriented medical paradigm