A Transdiagnostic Network Analysis of Childhood Trauma and Psychopathology.

BACKGROUND AND HYPOTHESIS: Psychiatric comorbidities suggest that symptoms overlap across different diagnoses; the transdiagnostic network approach is valuable for studying psychopathology. Childhood trauma is a common transdiagnostic risk factor for psychiatric disorders, but the complex relationship between childhood trauma and psychopathology has seldom been investigated using a large cross-sectional transdiagnostic sample. STUDY DESIGN: This study recruited 869 patients with different diagnoses, including 418 schizophrenia, 215 bipolar disorder, and 236 major depressive disorder. Participants completed psychiatric interviews and self-report questionnaires. We constructed dimension- and item-level Least Absolute Shrinkage and Selection Operator-based (LASSO) networks to explore the relationship between childhood trauma, psychopathology, and duration of illness. Moreover, we constructed directed acyclic graphs (DAGs) to tentatively clarify the potential directions of associations among these variables. Network Comparison Tests (NCTs) were conducted for different diagnostic groups and gender-stratified groups. STUDY RESULTS: The transdiagnostic LASSO networks showed that different types of childhood trauma exerted distinct impacts on various psychopathological dimensions. Emotional abuse was linked to depressive symptoms, physical abuse to excited symptoms, sexual abuse to positive and disorganized symptoms, emotional neglect to depressive symptoms and motivation and pleasure (MAP) deficits factor of negative symptoms, and physical neglect to MAP factor. The DAG findings generally concurred with the LASSO network. The NCT showed comparable networks. CONCLUSIONS: Our findings suggest that childhood trauma is significantly associated with the development of psychopathology across different diagnostic groups. The affective pathway model suggests that early identification and tailored interventions would be needed for people with a history of childhood trauma.

Revisiting the latent structure of negative symptoms in schizophrenia: Evidence from two second-generation clinical assessments.

BACKGROUND: Negative symptoms are core symptom of schizophrenia, and many previous research studied the latent structure of negative symptoms based on a single measurement scale. Applying two second-generation negative symptom scales to the same sample can address measurement-invariance of latent structure. METHODS: Three-hundred-and-five schizophrenia patients were assessed using the CAINS and the BNSS. Confirmatory Factor Analysis (CFA) tested four competing factor-models: (1) a 1-factor model; (2) a 2-factor model comprising the motivation and pleasure (MAP) domain and the diminished expression (EXP) domain; (3) a 5-factor model comprising anhedonia, avolition, asociality, blunted affect and alogia; (4) a hierarchical model comprising the "first-order" 5-domain factors and the "second-order" MAP & EXP factors. RESULTS: The CFA results for the data of the CAINS showed that the 2-factor model had the best data fit over the other competing models. The CFA using the BNSS data in the same sample also supported the superiority of the 2-factor model. Lastly, after combining the items of the BNSS and CAINS together in the same sample for CFA, the 2-factor model prevailed over the other competing models. CONCLUSIONS: The 2-factor model appears to be measurement-invariant latent structure of negative symptoms. The novel method of combining the items of the CAINS and BNSS might have circumvented the possible imperfect construct of a single scale. Our findings support the MAP and EXP factors as the latent structure for negative symptoms.

TKRD: Trusted kernel rootkit detection for cybersecurity of VMs based on machine learning and memory forensic analysis.

The promotion of cloud computing makes the virtual machine (VM) increasingly a target of malware attacks in cybersecurity such as those by kernel rootkits. Memory forensic, which observes the malicious tracks from the memory aspect, is a useful way for malware detection. In this paper, we propose a novel TKRD method to automatically detect kernel rootkits in VMs from private cloud, by combining VM memory forensic analysis with bio-inspired machine learning technology. Malicious features are extracted from the memory dumps of the VM through memory forensic analysis method. Based on these features, various machine learning classifiers are trained including Decision tree, Rule based classifiers, Bayesian and Support vector machines (SVM). The experiment results show that the Random Forest classifier has the best performance which can effectively detect unknown kernel rootkits with an Accuracy of 0.986 and an AUC value (the area under the receiver operating characteristic curve) of 0.998.