A major challenge today is personalizing the treatment for Major Depressive Disorder (MDD) patients in order to make it more efficient. In order to address this issue, we have proposed a novel approach based on machine learning models that utilize neural activity flow prior to treatment with selective serotonin reuptake inhibitor (SSRI) medication. The electroencephalogram (EEG) signals of 30 patients were used to calculate the neural activity flow of each patient based on the direct Directed Transfer Function (dDTF). Then, based on the area under the curve (AUC) values, 30 important connections were identified for delta, theta, alpha, beta, and gamma bands. In order to select the most important neural activity flow, these neural activity flows are combined, and forward features, mRMR, and Relief-F methods are applied. Lastly, Support vector machines (SVMs), decision tree, and random forest models are trained using selected neural activity flows. Results showed that the most connections came from F8, Pz, T5, and P4, which are mostly from the frontal and parietal lobes. In addition, the SVM model showed 98% accuracy in classification using forward feature selection, with most of the neural activity flows selected from alpha and beta. Finally, results indicate that patients who responded to treatment differed in their patterns of frontoparietal neural activity flows, which implies the Frontoparietal Network is primarily involved in treatment response at alpha and beta frequencies. Therefore, the proposed method is capable of accurately detecting responders in MDD patients, which can reduce costs for both patients and medical facilities.
نوع مطالعه:
Original |
موضوع مقاله:
Cognitive Neuroscience دریافت: 1402/3/18 | پذیرش: 1402/11/1