Abnormal Fractional Amplitude of Low-Frequency Fluctuation as a Potential Imaging Biomarker for First-Episode Major Depressive Disorder: A Resting-State fMRI Study and Support Vector Machine Analysis.

Yujun Gao,Yafen Wei,Hongwei Ren,Xi Wang,Zhenying Xiong,Dongbin Li,Ruoshi Liu

doi:10.3389/fneur.2021.751400

Abstract

Objective: Major depressive disorder (MDD) is a psychiatric disorder with serious negative health outcomes; however, there is no reliable method of diagnosis. This study explored the clinical diagnostic value of the fractional amplitude of low-frequency fluctuation (fALFF) based on the support vector machine (SVM) method for the diagnosis of MDD.Methods: A total of 198 first-episode MDD patients and 234 healthy controls were involved in this study, and all participants underwent resting-state functional magnetic resonance imaging (fMRI) scanning. Imaging data were analyzed with the fALFF and SVM methods.Results: Compared with the healthy controls, the first-episode MDD patients showed higher fALFF in the left mid cingulum, right precuneus, and left superior frontal gyrus (SFG). The increased fALFF in these three brain regions was positively correlated with the executive control reaction time (ECRT), and the increased fALFF in the left mid cingulum and left SFG was positively correlated with the 17-item Hamilton Rating Scale for Depression (HRSD-17) scores. The SVM results showed that increased fALFF in the left mid cingulum, right precuneus, and left SFG exhibited high diagnostic accuracy of 72.92% (315/432), 71.76% (310/432), and 73.84% (319/432), respectively. The highest diagnostic accuracy of 76.39% (330/432) was demonstrated for the combination of increased fALFF in the right precuneus and left SFG, along with a sensitivity of 84.34% (167/198), and a specificity of 70.51% (165/234).Conclusion: Increased fALFF in the left mid cingulum, right precuneus, and left SFG may serve as a neuroimaging marker for first-episode MDD. The use of the increased fALFF in the right precuneus and left SFG in combination showed the best diagnostic value.

Highlights

Major depressive disorder (MDD) is a severe psychiatric disorder affecting more than 264 million people around the world [1] and is characterized by cognitive and affective dysfunction
We investigated fractional amplitude of low-frequency fluctuation (fALFF) in first-episode MDD patients, screened brain regions that showed altered fALFF, and discussed their value as potential neuroimaging markers through the support vector machine (SVM) method
The increased fALFF values of these three brain regions in the firstepisode MDD patients were analyzed by the SVM method

Summary

Introduction

Major depressive disorder (MDD) is a severe psychiatric disorder affecting more than 264 million people around the world [1] and is characterized by cognitive and affective dysfunction. Previous studies have revealed that MDD participates in some mechanisms/pathophysiologies, such as genetics, environmental factors, neuroendocrinology, inflammation, neuroplasticity, and monoamines [2]. All these molecular and cellular alterations result in brain network structure and functional network connectivity. Several studies have confirmed that MDD is relevant to specific brain circuits or networks, such as the default mode network (DMN) [3], affective-salience circuit [4], and frontoparietal cognitive control circuit [5]. Major depressive disorder is related to activity within the subcallosal cingulate cortex, an extensively connected element of the limbic system that regulates feelings of sadness [7]. Major depressive disorder was associated with decreased connectivity between the ventromedial prefrontal cortex and dorsal striatum, which correlated with motor speed and psychomotor slowing [8]

Methods

Results

Discussion

Conclusion