The cerebral cortex and the cerebellum are spatially remote areas that are connected by complex circuits that link both primary and associative areas. Previous studies have revealed abnormalities in autism spectrum disorder (ASD); however, it is not clear whether cortico-cerebellar connectivity is differentially manifested in the disorder. To explore this issue, we investigated differences in intrinsic cortico-cerebellar functional connectivity between individuals with typical development (TD) and those with ASD. To this end, we used functional magnetic resonance imaging (fMRI) of 708 subjects under a resting state protocol provided by the ABIDE I Consortium. We found that people with ASD had diminished functional connectivity between the cerebellum and the following cortical regions: (i) right fusiform gyrus, (ii) right postcentral gyrus, (iii) right superior temporal gyrus, (iv) right middle temporal gyrus, and (v) left middle temporal gyrus. All of these regions are involved in many cognitive systems that contribute to commonly affected functions in ASD. For right fusiform gyrus, right superior temporal gyrus, and left middle temporal gyrus, we reproduced the results in an independent cohort composed of 585 subjects of the ABIDE II Consortium. Our results points toward a consistent atypical cortico-cerebellar connectivity in ASD.

No abstract

IntroductionClustering is usually the first exploratory analysis step in empirical data. When the data set comprises graphs, the most common approaches focus on clustering its vertices. In this work, we are interested in grouping networks with similar connectivity structures together instead of grouping vertices of the graph. We could apply this approach to functional brain networks (FBNs) for identifying subgroups of people presenting similar functional connectivity, such as studying a mental disorder. The main problem is that real-world networks present natural fluctuations, which we should consider.MethodsIn this context, spectral density is an exciting feature because graphs generated by different models present distinct spectral densities, thus presenting different connectivity structures. We introduce two clustering methods: k-means for graphs of the same size and gCEM, a model-based approach for graphs of different sizes. We evaluated their performance in toy models. Finally, we applied them to FBNs of monkeys under anesthesia and a dataset of chemical compounds.ResultsWe show that our methods work well in both toy models and real-world data. They present good results for clustering graphs presenting different connectivity structures even when they present the same number of edges, vertices, and degree of centrality.DiscussionWe recommend using k-means-based clustering for graphs when graphs present the same number of vertices and the gCEM method when graphs present a different number of vertices.

Background The present study aimed to apply multivariate pattern recognition methods to predict PTSD symptoms from whole-brain activation patterns during two contexts where the aversiveness of unpleasant pictures was manipulated by the presence or absence of safety cues. Methods Trauma-exposed participants were presented with neutral and mutilation pictures during fMRI collection. Before the presentation of pictures, a text informed the subjects that the pictures were fictitious (“safe context”) or real-life scenes (“real context”). We trained machine learning regression models (Gaussian process regression (GPR)) to predict PTSD symptoms in real and safe contexts. Results The GPR model could predict PTSD symptoms from brain responses to mutilation pictures in a real context but not a safe one. The brain regions with the highest contribution to the model were the occipito-parietal regions, including the superior parietal gyrus, inferior parietal gyrus, and supramarginal gyrus. Additional analysis showed that GPR regression models accurately predicted clusters of PTSD symptoms, nominally intrusion, avoidance, and alteration in cognition. As expected, we obtained very similar results as those obtained in a model predicting total PTSD symptoms. Conclusion These results are innovative by showing that machine learning applied to fMRI can predict not only PTSD total symptoms but also clusters of PTSD symptoms in a more aversive context. Furthermore, this approach was able to identify potential biomarkers for PTSD, especially in occipito-parietal regions.

Taiane Coelho Ramos. Técnicas de clusterização e estrati cação de indivíduos para estudo de redes funcionais cerebrais. Tese (Doutorado).