Less reduced gray matter volume in the subregions of superior temporal gyrus predicts better treatment efficacy in drug-naive, first-episode schizophrenia

Machine learning methods to predict outcomes of pharmacological treatment in psychosis

In recent years, machine learning (ML) has been a promising approach in the research of treatment outcome prediction in psychosis. In this study, we reviewed ML studies using different neuroimaging, neurophysiological, genetic, and clinical features to predict antipsychotic treatment outcomes in patients at different stages of schizophrenia. Literature available on PubMed until March 2022 was reviewed. Overall, 28 studies were included, among them 23 using a single-modality approach and 5 combining data from multiple modalities. The majority of included studies considered structural and functional neuroimaging biomarkers as predictive features used in ML models. Specifically, functional magnetic resonance imaging (fMRI) features contributed to antipsychotic treatment response prediction of psychosis with good accuracies. Additionally, several studies found that ML models based on clinical features might present adequate predictive ability. Importantly, by examining the additive effects of combining features, the predictive value might be improved by applying multimodal ML approaches. However, most of the included studies presented several limitations, such as small sample sizes and a lack of replication tests. Moreover, considerable clinical and analytical heterogeneity among included studies posed a challenge in synthesizing findings and generating robust overall conclusions. Despite the complexity and heterogeneity of methodology, prognostic features, clinical presentation, and treatment approaches, studies included in this review suggest that ML tools may have the potential to predict treatment outcomes of psychosis accurately. Future studies need to focus on refining feature characterization, validating prediction models, and evaluate their translation in real-world clinical practice.

The genetic influence of the DRD3 rs6280 polymorphism (Ser9Gly) on functional connectivity and gray matter volume of the hippocampus in patients with first-episode, drug-naïve schizophrenia

The D3 dopamine receptor (DRD3) plays a major role in cognitive function and is a candidate gene for schizophrenia. DRD3 is widely distributed in the hippocampus, but whether there are potential associations between the rs6280 genotype, the hippocampus, and cognitive function in first-episode, drug-naïve (FES) patients and healthy controls (HCs) is still poorly understood. First, using functional and structural magnetic resonance imaging data, we calculated the gray matter volume (GMV) and functional connectivity (FC) of the hippocampus. Then, we examined the possible interaction effect of the DRD3 genotype and the disease on the FC and GMV of the hippocampus in 52 FES patients and 51 HCs. Finally, the correlation between the FC and GMV in the hippocampus, influenced by rs6280, and the cognitive performance of subjects was analyzed. A significant interaction effect of diagnostic group by genotype of rs6280 on the GMV of the left hippocampus was found, with lower GMV in FES patients that were C carriers compared with TT homozygotes; the opposite pattern was found in the genetic subgroups of HCs. In the FES group, C carriers performed significantly worse on reasoning and problem-solving tests than TT homozygotes. The left hippocampal GMV positively correlated with reasoning and problem-solving performance in TT homozygotes, but this correlation disappeared in FES patients that were C carriers and in genetic subgroups of HCs. Together, these results suggest that FES patients that are C carriers of rs6280 have lower GMV in the hippocampus, resulting in greater cognitive impairment.

Decreased degree centrality values as a potential neuroimaging biomarker for migraine: A resting-state functional magnetic resonance imaging study and support vector machine analysis

Objective

Misdiagnosis and missed diagnosis of migraine are common in clinical practice. Currently, the pathophysiological mechanism of migraine is not completely known, and its imaging pathological mechanism has rarely been reported. In this study, functional magnetic resonance imaging (fMRI) technology combined with a support vector machine (SVM) was employed to study the imaging pathological mechanism of migraine to improve the diagnostic accuracy of migraine.

Methods

We randomly recruited 28 migraine patients from Taihe Hospital. In addition, 27 healthy controls were randomly recruited through advertisements. All patients had undergone the Migraine Disability Assessment (MIDAS), Headache Impact Test - 6 (HIT-6), and 15 min magnetic resonance scanning. We ran DPABI (RRID: SCR_010501) on MATLAB (RRID: SCR_001622) to preprocess the data and used REST (RRID: SCR_009641) to calculate the degree centrality (DC) value of the brain region and SVM (RRID: SCR_010243) to classify the data.

Results

Compared with the healthy controls (HCs), the DC value of bilateral inferior temporal gyrus (ITG) in patients with migraine was significantly lower and that of left ITG showed a positive linear correlation with MIDAS scores. The SVM results showed that the DC value of left ITG has the potential to be a diagnostic biomarker for imaging, with the highest diagnostic accuracy, sensitivity, and specificity for patients with migraine of 81.82, 85.71, and 77.78%, respectively.

Conclusion

Our findings demonstrate abnormal DC values in the bilateral ITG among patients with migraine, and the present results provide insights into the neural mechanism of migraines. The abnormal DC values can be used as a potential neuroimaging biomarker for the diagnosis of migraine.

Neural substrates of reward anticipation and outcome in schizophrenia: a meta-analysis of fMRI findings in the monetary incentive delay task

Dysfunction of the mesocorticolimbic dopaminergic reward system is a core feature of schizophrenia (SZ), yet its precise contributions to different stages of reward processing and their relevance to disease symptomology are not fully understood. We performed a coordinate-based meta-analysis, using the monetary incentive delay task, to identify which brain regions are implicated in different reward phases in functional magnetic resonance imaging in SZ. A total of 17 studies (368 SZ and 428 controls) were included in the reward anticipation, and 10 studies (229 SZ and 281 controls) were included in the reward outcome. Our meta-analysis revealed that during anticipation, patients showed hypoactivation in the striatum, anterior cingulate cortex, median cingulate cortex (MCC), amygdala, precentral gyrus, and superior temporal gyrus compared with controls. Striatum hypoactivation was negatively associated with negative symptoms and positively associated with the proportion of second-generation antipsychotic users (percentage of SGA users). During outcome, patients displayed hyperactivation in the striatum, insula, amygdala, hippocampus, parahippocampal gyrus, cerebellum, postcentral gyrus, and MCC, and hypoactivation in the dorsolateral prefrontal cortex (DLPFC) and medial prefrontal cortex (mPFC). Hypoactivity of mPFC during outcome was negatively associated with positive symptoms. Moderator analysis showed that the percentage of SGA users was a significant moderator of the association between symptom severity and brain activity in both the anticipation and outcome stages. Our findings identified the neural substrates for different reward phases in SZ and may help explain the neuropathological mechanisms underlying reward processing deficits in the disorder.