Investigating the predictive value of different resting-state functional MRI parameters in obsessive-compulsive disorder

Abstract

Previous resting-state functional magnetic resonance imaging (rs-fMRI) studies of obsessive-compulsive disorder (OCD) have facilitated our understanding of OCD pathophysiology based on its intrinsic activity. However, whether the group difference derived from univariate analysis could be useful for informing the diagnosis of individual OCD patients remains unclear. We aimed to apply multivariate pattern analysis of different rs-fMRI parameters to distinguish drug-naive patients with OCD from healthy control subjects (HCS). Fifty-four drug-naive OCD patients and 54 well-matched HCS were recruited. Four different rs-fMRI parameter maps, including the amplitude of low-frequency fluctuations (ALFF), fractional amplitude of low-frequency fluctuations (fALFF), regional homogeneity (ReHo) and functional connectivity strength (FCS), were calculated. Training of a support vector machine (SVM) classifier using rs-fMRI maps produced voxelwise discrimination maps. Overall, the classification accuracies were acceptable for the four rs-fMRI parameters. Excellent performance was achieved when ALFF maps were employed (accuracy, 95.37%, p < 0.01), good performance was achieved by using ReHo maps, weaker performance was achieved by using fALFF maps, and fair performance was achieved by using FCS maps. The brain regions showing the greatest discriminative power included the prefrontal cortex, anterior cingulate cortex, precentral gyrus, and occipital lobes. The application of SVM to rs-fMRI features may provide potential power for OCD classification.

Altered Global Brain Functional Connectivity in Drug-Naive Patients With Obsessive-Compulsive Disorder

“…Meanwhile, the activation in brain regions within the sensorimotor network in the inhibitory control processes may explain the essence of inhibitory control deficits of OCD (38). Furthermore, a recent study revealed that the amplitude of low-frequency fluctuations of the precentral gyrus showed a large discriminative power to identify patients with OCD from HCs (39).…”

Decreased Resting-State Interhemispheric Functional Connectivity in Medication-Free Obsessive-Compulsive Disorder

“…Based on our current and previous findings, we inferred that decreased ReHo and interhemispheric functional homotopy in the occipital cortex and increased FC with the CSTC circuits may work together to contribute to the pathogenesis of OCD. The precentral and postcentral gyri are important brain regions in the sensorimotor network; decreased global brain FC in the precentral or postcentral gyrus at rest has been found in patients with OCD ( 39 ), and the amplitude of the low-frequency fluctuations of the precentral gyrus can be used to distinguish patients with OCD from HCs ( 8 ). In the present study, decreased interhemispheric functional homotopy in the precentral and postcentral gyri may be associated with damaged sensory-motor integration and sensory gating in OCD at rest and might contribute to the inability to suppress internally repetitive and intrusive thoughts and behaviors in patients with OCD ( 40 , 41 ).…”

“…In the present study, decreased interhemispheric functional homotopy in the precentral and postcentral gyri may be associated with damaged sensory-motor integration and sensory gating in OCD at rest and might contribute to the inability to suppress internally repetitive and intrusive thoughts and behaviors in patients with OCD ( 40 , 41 ). Brain regions with macro- and microstructural alterations outside the CSTC circuits are also related to the pathophysiology of OCD ( 6 – 8 , 11 , 42 – 44 ). The present findings provide additional evidence to elucidate the pathogenesis of OCD.…”

Advances in multimodal data fusion in neuroimaging: Overview, challenges, and novel orientation

“…fMRI studies can provide information about the pathophysiology of OCD [164, 165] . However, whether this information from single fMRI modality alone could be of clinical value in the diagnosis of individual patients is not clear [166] . The pathophysiological feature of OCD, as revealed by fMRI studies, suggest that abnormal brain metabolites may be implied in OCDs.…”