Exploring the use of electroencephalogram signals for medical diagnosis

Main Article Content

Ms. Pranjali Deepak Nikam, Dr. Prashantha G. R., Ms. Sonali Mahendra Sonavane


A vast grouping of EEG portrayal models is put forth by research scholars all through the long haul, and everyone vary identical to congruity, exactness, survey, accuracy and concede execution. For instance, work in [2, 3, 4] looks at plan of lessened direction set (RISC)- V convolutional Neural Network (CNN) Coprocessor, KNN (k Nearest Neighbor), blend of direct discriminant examination (LDA), ANN (counterfeit brain organization) and support vector machine (SVM) with ordinary spatial model (CSP), and Transfer TSK Fuzzy Classifier (TTFC) for achieving better portrayal results. These models have extraordinary precision, yet need terms of exactness execution in light of their application-unequivocal portrayal characteristics. Developments to this model are discussed in [5, 6], wherein LIFUS (low-force trotted ultrasound energy) and NNM (Neuroglial Network Model) are used for multidomain EEG groupings. They have extraordinary exactness, yet can't be scaled for a long while in light of high computational unpredictability. To vanquish the above issues, next region proposes wavelet pressure based quadratic model for EEG request utilizing multivariate examination that helps high-capability and high flexibility EEG portrayal for various clinical circumstances.

Article Details