No abstract

No abstract

It is unclear whether abnormal spontaneous neural activation patterns found in chronic schizophrenia patients (CSP) are part of the pathogenesis of disease, consequences of chronic illness, or effects of antipsychotic treatment. We performed a longitudinal resting-state functional magnetic resonance imaging (fMRI) study in 42 treatment-naïve first-episode schizophrenia patients (FESP) at baseline and then after 8-weeks of risperidone monotherapy, and compared the findings to 38 healthy volunteers. Spontaneous brain activity was quantified using the fractional amplitude of low-frequency fluctuations (fALFF) and regional homogeneity (ReHo) and compared between patients and controls. Pretreatment, patients exhibited higher fALFF in left caudate compared with controls. After treatment, patients had elevated fALFF in bilateral putamen and right caudate, and increased ReHo in right caudate and left putamen. Greater increase of fALFF in the left putamen correlated with less improvement in positive symptoms. Thus, abnormalities of spontaneous neural activity in chronic schizophrenia is at least partly due to a medication effect. The observed post-treatment increase in striatal intrinsic activity may reflect counter-therapeutic functional adaptation to dopamine D2 receptor occupancy required for medication effects on psychosis.

Predictive coding has been widely used in legal matters to find relevant or privileged documents in large sets of electronically stored information. It saves the time and cost significantly. Logistic Regression (LR) and Support Vector Machines (SVM) are two popular machine learning algorithms used in predictive coding. Recently, deep learning received a lot of attentions in many industries. This paper reports our preliminary studies in using deep learning in legal document review. Specifically, we conducted experiments to compare deep learning results with results obtained using a SVM algorithm on the four datasets of real legal matters. Our results showed that CNN performed better with larger volume of training dataset and should be a fit method in the text classification in legal industry.