Identification of male schizophrenia patients using brain morphology based on machine learning algorithms.

BACKGROUND: Schizophrenia is a severe psychiatric disease, and its prevalence is higher. However, diagnosis of early-stage schizophrenia is still considered a challenging task. AIM: To employ brain morphological features and machine learning method to differentiate male individuals with schizophrenia from healthy controls. METHODS: The least absolute shrinkage and selection operator and t tests were applied to select important features from structural magnetic resonance images as input features for classification. Four commonly used machine learning algorithms, the general linear model, random forest (RF), k-nearest neighbors, and support vector machine algorithms, were used to develop the classification models. The performance of the classification models was evaluated according to the area under the receiver operating characteristic curve (AUC). RESULTS: A total of 8 important features with significant differences between groups were considered as input features for the establishment of classification models based on the four machine learning algorithms. Compared to other machine learning algorithms, RF yielded better performance in the discrimination of male schizophrenic individuals from healthy controls, with an AUC of 0.886. CONCLUSION: Our research suggests that brain morphological features can be used to improve the early diagnosis of schizophrenia in male patients.

Potential role between inflammatory cytokines and Tie-2 receptor levels and clinical symptoms in patients with first-episode schizophrenia.

BACKGROUND: Schizophrenia (SCZ) is associated with chronic low-grade inflammation, which may be involved in the underlying pathological mechanism of the disease and may influence patient prognosis. We evaluated the differences in serum cytokine and Tie-2 receptor levels between patients with first-episode SCZ and healthy controls and explored the correlation thereof with clinical symptoms. METHODS: Seventy-six participants were recruited for the present study, including 40 patients with first-episode SCZ and 36 healthy controls. Positive and Negative Syndrome Scale (PANSS) and Brief Psychiatric Rating Scale (BPRS) scores, demographic data, and blood samples were collected at baseline. A hypersensitive Meso Scale Discovery (MSD) electrochemiluminescence assay system was used to measure cytokine and Tie-2 receptor levels. Spearman's correlation and stepwise linear regression were used to analyze the data. RESULTS: Serum interleukin-1ß and -4 levels were significantly increased, and Tie-2 levels were significantly decreased, in first-episode SCZ patients as compared to healthy controls. IL-1ß levels were positively correlated with total BPRS scores, resistance subscores, and PANSS positive subscores. Furthermore, IL-1ß levels were negatively correlated with Tie-2 receptor expression levels. Stepwise linear regression analysis demonstrated that IL-1ß levels correlated positively with PANSS positive subscores and BPRS total scores. PANSS negative subscores, general psychopathology subscores, and PANSS total scores had positive effects on the Tie-2 receptor. Receiver operating characteristic (ROC) curve analysis showed that IL-1ß and Tie-2 were highly sensitive and specific for predicting first-episode SCZ symptoms and achieving an area under the ROC curve of 0.8361 and 0.6462, respectively. CONCLUSION: Our results showed that patients with first-episode SCZ have low-grade inflammation. IL-1ß and Tie-2 receptors may be important mediators between inflammation and vascular dysfunction in patients with SCZ and may underlie the increased cardiovascular disease in this population. TRIAL REGISTRATION: The clinical trial registration date was 06/11/2018, registration number was chiCTR1800019343.

Correlation between elevated serum interleukin-1ß, interleukin-16 levels and psychiatric symptoms in patients with schizophrenia at different stages.

BACKGROUND: There is increasing evidence that immune dysfunction plays an important role in the pathogenesis of schizophrenia. Meso Scale Discovery (MSD) is bioanalytical method, which can detect serum inflammatory factors in patients. MSD has higher sensitivities, capturing a narrower range of proteins compared to other methods typically used in similar studies. The present study was aimed to explore the correlation between the levels of serum inflammatory factors and psychiatric symptoms in patients with schizophrenia at different stages and investigate a wide panel of inflammatory factors as independent factors for the pathogenesis of schizophrenia. METHODS: We recruited 116 participants, including patients with first-episode schizophrenia (FEG, n = 40), recurrence patients (REG, n = 40) with relapse-episode schizophrenia, and a control group (healthy people, HP, n = 36). Patients are diagnosed according to the DSM -V. The plasma levels of IFN-γ, IL-10, IL-1ß, IL-2, IL-6, TNF-α, CRP, VEGF, IL-15, and IL-16 were tested by the MSD technique. Patient-related data was collected, including sociodemographic data, positive and negative symptom scale (PANSS), and brief psychiatric rating scale (BPRS) and subscale scores. The independent sample T test, χ2 test, Analysis of covariance (ANCOVA), the least significant difference method (LSD), Spearman's correlation test, binary logistic regression analysis and ROC curve analysis were used in this study. RESULTS: There were significant differences in serum IL-1ß (F = 2.37, P = 0.014) and IL-16 (F = 4.40, P < 0.001) levels among the three groups. The level of serum IL-1ß in the first-episode group was significantly higher than in the recurrence group (F = 0.87, P = 0.021) and control group (F = 2.03, P = 0.013), but there was no significant difference between the recurrence group and control group (F = 1.65, P = 0.806). The serum IL-16 levels in the first-episode group (F = 1.18, P < 0.001) and the recurrence group (F = 0.83, P < 0.001) were significantly higher than in the control group, and there was no significant difference between the first-episode group and the recurrence group (F = 1.65, P = 0.61). Serum IL-1ß was negatively correlated with the general psychopathological score (GPS) of PANSS (R=-0.353, P = 0.026). In the recurrence group, serum IL-16 was positively correlated with the negative score (NEG) of the PANSS scale (R = 0.335, P = 0.035) and negatively correlated with the composite score (COM) (R=-0.329, P = 0.038). In the study, IL-16 levels were an independent variable of the onset of schizophrenia both in the first-episode (OR = 1.034, P = 0.002) and recurrence groups (OR = 1.049, P = 0.003). ROC curve analysis showed that the areas under IL-16(FEG) and IL-16(REG) curves were 0.883 (95%CI:0.794-0.942) and 0.887 (95%CI:0.801-0.950). CONCLUSIONS: Serum IL-1ß and IL-16 levels were different between patients with schizophrenia and healthy people. Serum IL-1ß levels in first-episode schizophrenia and serum IL-16 levels in relapsing schizophrenia were correlated with the parts of psychiatric symptoms. The IL-16 level may be an independent factor associating with the onset of schizophrenia.

Prediction of violence in male schizophrenia using sMRI, based on machine learning algorithms.

BACKGROUND: Violent behavior in patients with schizophrenia (SCZ) is a major social problem. The early identification of SCZ patients with violence can facilitate implementation of targeted intervention. METHODS: A total of 57 male SCZ patients were recruited into this study. The general linear model was utilized to compare differences in structural magnetic resonance imaging (sMRI) including gray matter volume, cortical surface area, and cortical thickness between 30 SCZ patients who had exhibited violence and 27 SCZ patients without a history of violence. Based on machine learning algorithms, the different sMRI features between groups were integrated into the models for prediction of violence in SCZ patients. RESULTS: After controlling for the whole brain volume and age, the general linear model showed significant reductions in right bankssts thickness, inferior parietal thickness as well as left frontal pole volume in the patients with SCZ and violence relative to those without violence. Among seven machine learning algorithms, Support Vector Machine (SVM) have better performance in differentiating patients with violence from those without violence, with its balanced accuracy and area under curve (AUC) reaching 0.8231 and 0.841, respectively. CONCLUSIONS: Patients with SCZ who had a history of violence displayed reduced cortical thickness and volume in several brain regions. Based on machine learning algorithms, structural MRI features are useful to improve predictive ability of SCZ patients at particular risk of violence.

Childhood maltreatment increases the suicidal risk in Chinese schizophrenia patients.

Objectives: Childhood trauma might be a modifiable risk factor among adults with serious mental illness. However, the correlation of child trauma and suicide is unclear, which were cited most frequently as the biggest challenge to schizophrenia (SCZ) patients in China. We aim to study relationships between child trauma and suicide in SCZ patients of different disease stages. Methods: Ninety-one participants were included and divided into two groups, namely, first-episode group (n = 46), relapsed group (n = 45). The Positive and Negative Syndrome Scale was used to evaluate the severity of psychotic symptoms. The Beck's Suicide Intent Scale and The Nurses' Global Assessment of Suicide Risk were conducted by patient self-report to assess suicide symptom. The childhood trauma questionnaire was used to estimate severity of traumatic stress experienced during childhood. Results: Childhood trauma and different dimensions of suicide were significantly higher in the relapsed group than first-episode group (P < 0.01, respectively). BMI has a significant positive relationship with recent psychosocial stress (ß = 0.473, t = 3.521, P < 0.001) in first-episode group. As in relapsed group, BMI has a positive effect between severe mental illness and suicide ideation (ß = 0.672, t = 5.949, P < 0.001; ß = 0.909, t = 2.463, P < 0.001), Furthermore, emotional neglect presented positively related to the suicide risk and proneness to suicidal behavior (ß = 0.618, t = 5.518, P < 0.001; ß = 0.809, t = 5.356, P < 0.001). Conclusion: Relapsed group of patients had significantly more severe childhood trauma, recent psychosocial stress, suicidal risk and proneness to suicidal behavior. BMI and emotional neglect are unique predictors for different dimensions of suicide.