Prolonged Disease Course Leads to Impaired Brain Function in Anxiety Disorder: A Resting State EEG Study

Abstract

Shaoqi Mou,1,* Shiyu Yan,1,* Shanhong Shen,2 Yibin Shuai,2 Gang Li,3 Zhongxia Shen,2 Ping Shen2 1Department of Psychiatry, Wenzhou Medical University, Wenzhou, People’s Republic of China; 2Department of Sleep Medical Center, Huzhou Third Municipal Hospital, The Affiliated Hospital of Huzhou University, Huzhou, People’s Republic of China; 3College of Engineering, Zhejiang Normal University, Jinhua, People’s Republic of China*These authors contributed equally to this workCorrespondence: Zhongxia Shen; Ping Shen, Email [email protected]; [email protected]: Anxiety disorder (AD) is a common disabling disease. The prolonged disease course may lead to impaired cognitive performance, brain function, and a bad prognosis. Few studies have examined the effect of disease course on brain function by electroencephalogram (EEG).Methods: Resting-state EEG analysis was performed in 34 AD patients. The 34 patients with AD were divided into two groups according to the duration of their illness: anxious state (AS) and generalized anxiety disorder (GAD). Then, EEG features, including univariate power spectral density (PSD), fuzzy entropy (FE), and multivariable functional connectivity (FC), were extracted and compared between AS and GAD. These features were evaluated by three previously validated machine learning methods to test the accuracy of classification in AS and GAD.Results: Significant decreased PSD and FE in GAD were detected compared with AS, especially in the Alpha 2 band. In addition, FC analysis indicated that GAD patients’ connection between the left and right hemispheres decreased. Based on machine learning, AS and GAD are classified on a six-month criterion with the highest classification accuracy of up to 0.99 ± 0.0015.Conclusion: The brain function of patients is more severely impaired in AD patients with longer illness duration. Resting-state EEG demonstrated to be a promising examination in the classification in GAD and AS using machine learning methods with better classification accuracy.Keywords: electroencephalogram, EEG, disease course, anxiety disorder, brain function, machine learning

Keywords

• electroencephalogram
• eeg
• disease course
• anxiety disorder
• brain function
• machine learning.