Alterations in neural activation in the ventral frontoparietal network during complex magnocellular stimuli in developmental dyslexia associated with READ1 deletion.

BACKGROUND: An intronic deletion within intron 2 of the DCDC2 gene encompassing the entire READ1 (hereafter, READ1d) has been associated in both children with developmental dyslexia (DD) and typical readers (TRs), with interindividual variation in reading performance and motion perception as well as with structural and functional brain alterations. Visual motion perception -- specifically processed by the magnocellular (M) stream -- has been reported to be a solid and reliable endophenotype of DD. Hence, we predicted that READ1d should affect neural activations in brain regions sensitive to M stream demands as reading proficiency changes. METHODS: We investigated neural activations during two M-eliciting fMRI visual tasks (full-field sinusoidal gratings controlled for spatial and temporal frequencies and luminance contrast, and sensitivity to motion coherence at 6%, 15% and 40% dot coherence levels) in four subject groups: children with DD with/without READ1d, and TRs with/without READ1d. RESULTS: At the Bonferroni-corrected level of significance, reading skills showed a significant effect in the right polar frontal cortex during the full-field sinusoidal gratings-M task. Regardless of the presence/absence of the READ1d, subjects with poor reading proficiency showed hyperactivation in this region of interest (ROI) compared to subjects with better reading scores. Moreover, a significant interaction was found between READ1d and reading performance in the left frontal opercular area 4 during the 15% coherent motion sensitivity task. Among subjects with poor reading performance, neural activation in this ROI during this specific task was higher for subjects without READ1d than for READ1d carriers. The difference vanished as reading skills increased. CONCLUSIONS: Our findings showed a READ1d-moderated genetic vulnerability to alterations in neural activation in the ventral attentive and salient networks during the processing of relevant stimuli in subjects with poor reading proficiency.

Brain Alteration Patterns in Children with Duchenne Muscular Dystrophy: A Machine Learning Approach to Magnetic Resonance Imaging.

Background: Duchenne Muscular Dystrophy (DMD) is a genetic disease in which lack of the dystrophin protein causes progressive muscular weakness, cardiomyopathy and respiratory insufficiency. DMD is often associated with other cognitive and behavioral impairments, however the correlation of abnormal dystrophin expression in the central nervous system with brain structure and functioning remains still unclear. Objective: To investigate brain involvement in patients with DMD through a multimodal and multivariate approach accounting for potential comorbidities. Methods: We acquired T1-weighted and Diffusion Tensor Imaging data from 18 patients with DMD and 18 age- and sex-matched controls with similar cognitive and behavioral profiles. Cortical thickness, structure volume, fractional anisotropy and mean diffusivity measures were used in a multivariate analysis performed using a Support Vector Machine classifier accounting for potential comorbidities in patients and controls. Results: the classification experiment significantly discriminates between the two populations (97.2% accuracy) and the forward model weights showed that DMD mostly affects the microstructural integrity of long fiber bundles, in particular in the cerebellar peduncles (bilaterally), in the posterior thalamic radiation (bilaterally), in the fornix and in the medial lemniscus (bilaterally). We also reported a reduced cortical thickness, mainly in the motor cortex, cingulate cortex, hippocampal area and insula. Conclusions: Our study identified a small pattern of alterations in the CNS likely associated with the DMD diagnosis.

Review on deep learning fetal brain segmentation from Magnetic Resonance images.

Brain segmentation is often the first and most critical step in quantitative analysis of the brain for many clinical applications, including fetal imaging. Different aspects challenge the segmentation of the fetal brain in magnetic resonance imaging (MRI), such as the non-standard position of the fetus owing to his/her movements during the examination, rapid brain development, and the limited availability of imaging data. In recent years, several segmentation methods have been proposed for automatically partitioning the fetal brain from MR images. These algorithms aim to define regions of interest with different shapes and intensities, encompassing the entire brain, or isolating specific structures. Deep learning techniques, particularly convolutional neural networks (CNNs), have become a state-of-the-art approach in the field because they can provide reliable segmentation results over heterogeneous datasets. Here, we review the deep learning algorithms developed in the field of fetal brain segmentation and categorize them according to their target structures. Finally, we discuss the perceived research gaps in the literature of the fetal domain, suggesting possible future research directions that could impact the management of fetal MR images.