Altered cerebellar lobular volumes correlate with clinical deficits in siblings and children with ASD: evidence from toddlers.

BACKGROUND: Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder characterized by impaired social and communication skills, narrow interests, and repetitive behavior. It is known that the cerebellum plays a vital role in controlling movement and gait posture. However, recently, researchers have reported that the cerebellum may also be responsible for other functions, such as social cognition, reward, anxiety, language, and executive functions. METHODS: In this study, we ascertained volumetric differences from cerebellar lobular analysis from children with ASD, ASD siblings, and typically developing healthy controls. In this cross-sectional study, a total of 30 children were recruited, including children with ASD (N = 15; mean age = 27.67 ± 5.1 months), ASD siblings (N = 6; mean age = 17.5 ± 3.79 months), and typically developing children (N = 9; mean age = 17.67 ± 3.21 months). All the MRI data was acquired under natural sleep without using any sedative medication. We performed a correlation analysis with volumetric data and developmental and behavioral measures obtained from these children. Two-way ANOVA and Pearson correlation was performed for statistical data analysis. RESULTS: We observed intriguing findings from this study, including significantly increased gray matter lobular volumes in multiple cerebellar regions including; vermis, left and right lobule I-V, right CrusII, and right VIIb and VIIIb, respectively, in children with ASD, compared to typically developing healthy controls and ASD siblings. Multiple cerebellar lobular volumes were also significantly correlated with social quotient, cognition, language, and motor scores with children with ASD, ASD siblings, and healthy controls, respectively. CONCLUSIONS: This research finding helps us understand the neurobiology of ASD and ASD-siblings, and critically advances current knowledge about the cerebellar role in ASD. However, results need to be replicated for a larger cohort from longitudinal research study in future.

Performance of machine learning classification models of autism using resting-state fMRI is contingent on sample heterogeneity.

Autism spectrum disorders (ASDs) are heterogeneous neurodevelopmental conditions. In fMRI studies, including most machine learning studies seeking to distinguish ASD from typical developing (TD) samples, cohorts differing in gender and symptom severity composition are often treated statistically as one "ASD group". Using resting-state functional connectivity (FC) data, we implemented random forest to build diagnostic classifiers in 4 ASD samples including a total of 656 participants (NASD = 306, NTD = 350, ages 6-18). Groups were manipulated to titrate heterogeneity of gender and symptom severity and partially overlapped. Each sample differed on inclusionary criteria: (1) all genders, unrestricted severity range; (2) only male participants, unrestricted severity; (3) all genders, higher severity only; (4) only male participants, higher severity. Each set consisted of 200 participants per group (ASD, TD; matched on age and head motion), 160 for training and 40 for validation. FMRI time series from 237 regions of interest (ROIs) were Pearson correlated in a 237×237 FC matrix and classifiers were built using random forest in training samples. Classification accuracies in validation samples were 62.5%, 65%, 70% and 73.75%, respectively for samples 1-4. Connectivity within cingulo-opercular task control (COTC) network, and between COTC ROIs and default mode and dorsal attention network contributed overall most informative features, but features differed across sets. Findings suggest that diagnostic classifiers vary depending on ASD sample composition. Specifically, greater homogeneity of samples regarding gender and symptom severity enhances classifier performance. However, given the true heterogeneity of ASDs, performance metrics alone may not adequately reflect classifier utility.

Subcortical Region Segmentation using Fuzzy Based Augmented Lagrangian Multiphase Level Sets Method in Autistic MR Brain Images.

In this work, the subcortical regions of control and autistic MR brain are segmented from the skull stripped images using Fuzzy C-Means (FCM) based Augmented Lagrangian (AL) multiphase level set method. The FCM method is used as he intensity discriminator for the multiphase level set method. The AL function avoids the re-initialization procedure. The segmented subcortical regions are validated with the ground truth images using dice similarity index. The texture features such as energy and entropy are calculated from the extracted cortical and subcortical regions. The results show that the multiphase level set method is able to segment the subcortical regions such as corpus callosum, brain stem and cerebellum. The dice similarity index gives above 0.85 for controls and 0.8 for autistic subjects. The texture feature energy calculated from the cortical region is high in autistics compared to the control subjects and vice versa in the case of entropy. The energy calculated from the subcortical regions is high in controls and entropy is high in autism subjects. Comparatively, the energyand entropy calculated from the total brain and brain stem gives significant variation (p<0.0001) between the control and autistic images. As the delayed growth of subcortical region is associated with high values of entropy, this study is clinically significant in the mass screening of neurodevelopmental disorders such as autism.