2021/01/14 by Sajila Wickramaratne, Wickramaratne, Sajila D., Md Shaad Mahmud +1
Engineering · Medicine · Neuroscience · #EEG and Brain-Computer Interfaces #FOS: Computer and information sciences #Machine Learning (cs.LG) #Non-Invasive Vital Sign Monitoring #Optical Imaging and Spectroscopy Techniques
paper · pdf · doi:10.48550/arxiv.2101.05891
openalex publication_date 2021/01/14 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
Functional near-infrared spectroscopy (fNIRS) is a non-invasive, economical\nmethod used to study its blood flow pattern. These patterns can be used to\nclassify tasks a subject is performing. Currently, most of the classification\nsystems use simple machine learning solutions for the classification of tasks.\nThese conventional machine learning methods, which are easier to implement and\ninterpret, usually suffer from low accuracy and undergo a complex preprocessing\nphase before network training. The proposed method converts the raw fNIRS time\nseries data into an image using Gramian Angular Summation Field. A Deep\nConvolutional Neural Network (CNN) based architecture is then used for task\nclassification, including mental arithmetic, motor imagery, and idle state.\nFurther, this method can eliminate the feature selection stage, which affects\nthe traditional classifiers' performance. This system obtained 87.14% average\nclassification accuracy higher than any other method for the dataset.\n