Identification of voxel-based texture abnormalities as new biomarkers for schizophrenia and major depressive patients using layer-wise relevance propagation on deep learning decisions.

Non-segmented MRI brain images are used for the identification of new Magnetic Resonance Imaging (MRI) biomarkers able to differentiate between schizophrenic patients (SCZ), major depressive patients (MD) and healthy controls (HC). Brain texture measures such as entropy and contrast, capturing the neighboring variation of MRI voxel intensities, were computed and fed into deep learning technique for group classification. Layer-wise relevance was applied for the localization of the classification results. Texture feature map of non-segmented brain MRI scans were extracted from 141 SCZ, 103 MD and 238 HC. The gray level co-occurrence matrix (GLCM) was calculated on a voxel-by-voxel basis in a cube of voxels. Deep learning tested if texture feature map could predict diagnostic group membership of three classes under a binary classification (SCZ vs. HC, MD vs. HC, SCZ vs. MD). The method was applied in a repeated nested cross-validation scheme and cross-validated feature selection. The regions with the highest relevance (positive/negative) are presented. The method was applied on non-segmented images reducing the computation complexity and the error associated with segmentation process.

Serotonergic modulation of 'waiting impulsivity' is mediated by the impulsivity phenotype in humans.

In rodents, the five-choice serial reaction time task (5-CSRTT) has been established as a reliable measure of waiting impulsivity being defined as the ability to regulate a response in anticipation of reinforcement. Key brain structures are the nucleus accumbens (NAcc) and prefrontal regions (for example, pre- and infralimbic cortex), which are, together with other transmitters, modulated by serotonin. In this functional magnetic resonance imaging study, we examined 103 healthy males while performing the 5-CSRTT measuring brain activation in humans by means of a paradigm that has been widely applied in rodents. Subjects were genotyped for the tryptophan hydroxylase-2 (TPH2; G-703T; rs4570625) variant, an enzyme specific for brain serotonin synthesis. We addressed neural activation patterns of waiting impulsivity and the interaction between the NAcc and the ventromedial prefrontal cortex (vmPFC) using dynamic causal modeling. Genetic influence was examined via interaction analyses between the TPH2 genotype (GG homozygotes vs T allele carriers) and the degree of impulsivity as measured by the 5-CSRTT. We found that the driving input of the vmPFC was reduced in highly impulsive T allele carriers (reflecting a reduced top-down control) in combination with an enhanced response in the NAcc after correct target processing (reflecting an augmented response to monetary reward). Taken together, we found a high overlap of our findings with reports from animal studies in regard to the underlying cognitive processes, the brain regions associated with waiting impulsivity and the neural interplay between the NAcc and vmPFC. Therefore, we conclude that the 5-CSRTT is a promising tool for translational studies.