DATA TRANSFORMATION OF EOG- EEG SENSOR DEVICE AND LARGE-SCALE MACHINE LEARNING METHODS FOR IOT HEALTH-CARE SYSTEM IN DETECTING DIGITAL EYE-STRAIN SYNDROME

Home » Uncategorized » DATA TRANSFORMATION OF EOG- EEG SENSOR DEVICE AND LARGE-SCALE MACHINE LEARNING METHODS FOR IOT HEALTH-CARE SYSTEM IN DETECTING DIGITAL EYE-STRAIN SYNDROME

...to create

THE BEST SLIDER

with no compromises!

DATA TRANSFORMATION OF EOG- EEG SENSOR DEVICE AND LARGE-SCALE MACHINE LEARNING METHODS FOR IOT HEALTH-CARE SYSTEM IN DETECTING DIGITAL EYE-STRAIN SYNDROME

Dr. Norma Alias

Associate Professor and Researcher, University Teknologi Malaysia, Johor Bahru, Johor, Malaysia

normaalias@utm.my

Article

Worldwide, Sightsavers reported 36 million people are blind, but 75 percent of this sight loss can be cured or prevented. In the era of 4iR, our eyes are responsible for fourfifths of data collection our brain receives. Bonilla-Warford, 2020 mentioned digital eye-strain syndrome is a major problem for Americans. This syndrome can decrease work productivity because of eye strain, neck pain, headaches, and blurred vision. The signal filtering of EOG and EEG sensor devices able to predict the weakness of eye movement-blinks, muscle strain-stress, and brain waves affected by the syndrome. Based on the drawback, this research focuses on machine learning technology to analyze a big dataset generated by EOG- EEG sensor devices. The integration of Internet of Thing (IoT) architecture and its components increase the speedup of data transformation through the embedded sensor device. As an impact, the process of big data analytics of the health-care system will be more efficient. Thus, this research will highlight intelligent data processing and machine learning algorithms to detect the weakness of eye movement-blinks, muscle strain-stress, and brain waves. The most critical challenges related to big data simulations are high-performance computing (HPC) and its computational platform system.  Thus, large-scale machine learning required the transformation of sequential to the parallel algorithm, domain decomposition strategy, message passing paradigm, granularity indicator, and parallel performance evaluations. The performance metrics of machine learning technology is based on sensitivity, precision, specificity, and accuracy. The superior classifier will be selected to predict eye-movement,  brain wave analytics, and its variability at longer periods of time. This health-care system well suits to recognize and interpret eye fatigue, blurred vision, and abnormal brain wave to facilitate efficient interaction and personalization. The function of alerting to caretakers and informing hospital in critical condition will make this intelligent health-care system new and unique.

Relevant Links: www.utm.my 

2020-06-25T16:23:59+00:00
Translate »