REVIEW STUDY FOR MIGRAINE IDENTIFICATION USING BRAIN SIGNAL ANALYSIS
Migraine is a neurological disease characterized by pulsating pain on one side of the head. Sometimes it is accompanied by sensory and visual symptoms. The symptoms may worsen and turn from an occasional headache to a chronic headache. In some cases, it develops into other diseases such as epilepsy and stroke, which affects the health of the individual and society, causing the loss of many working hours. Therefore, the researchers proposed several ways to diagnose this disease ccombined with machine and deep learning techniques in order to advance the health situation and develop the medical care provided to the patient away from traditional methods that are usually cumbersome and take a long time. These methods include recording brain signals and analyzing them to extract distinctive features and benefit from them later in diagnosing and classifying the disease. As well as the MRI method) magnetic resonance imaging (of the brain to detect changes occurring in the functional connectivity of different brain regions associated with pain. So, the primary aim of this review was to evaluate the effectiveness of machine and deep learning algorithms to predict and detection migraine and its types from EEG and MRI method. The results show that deep learning has the ability to predict migraines based on EEG, which could eventually be used to assist in clinical care. Based on the resting state, CWT method, and AlexNet classifier, the scenario, approach, and model that demonstrated the greatest successful performance for early diagnosis of migraine (99.74%) were determined. We believe that the results of this study are encouraging and useful for early detection of migraines based on electroencephalogram (EEG)
Keywords: Migraine Identification, Brain Signal Analysis