Automated motion artifact detection in early pediatric diffusion MRI using a convolutional neural network.

Diffusion MRI (dMRI) is a widely used method to investigate the microstructure of the brain. Quality control (QC) of dMRI data is an important processing step that is performed prior to analysis using models such as diffusion tensor imaging (DTI) or neurite orientation dispersion and density imaging (NODDI). When processing dMRI data from infants and young children, where intra-scan motion is common, the identification and removal of motion artifacts is of the utmost importance. Manual QC of dMRI data is (1) time-consuming due to the large number of diffusion directions, (2) expensive, and (3) prone to subjective errors and observer variability. Prior techniques for automated dMRI QC have mostly been limited to adults or school-age children. Here, we propose a deep learning-based motion artifact detection tool for dMRI data acquired from infants and toddlers. The proposed framework uses a simple three-dimensional convolutional neural network (3DCNN) trained and tested on an early pediatric dataset of 2,276 dMRI volumes from 121 exams acquired at 1 month and 24 months of age. An average classification accuracy of 95% was achieved following four-fold cross-validation. A second dataset with different acquisition parameters and ages ranging from 2-36 months (consisting of 2,349 dMRI volumes from 26 exams) was used to test network generalizability, achieving 98% classification accuracy. Finally, to demonstrate the importance of motion artifact volume removal in a dMRI processing pipeline, the dMRI data were fit to the DTI and NODDI models and the parameter maps were compared with and without motion artifact removal.

Tract- and gray matter- based spatial statistics show white matter and gray matter microstructural differences in autistic males.

Background: Autism spectrum disorder (ASD) is a neurodevelopmental condition commonly studied in the context of early childhood. As ASD is a life-long condition, understanding the characteristics of brain microstructure from adolescence into adulthood and associations to clinical features is critical for improving outcomes across the lifespan. In the current work, we utilized Tract Based Spatial Statistics (TBSS) and Gray Matter Based Spatial Statistics (GBSS) to examine the white matter (WM) and gray matter (GM) microstructure in neurotypical (NT) and autistic males. Methods: Multi-shell diffusion MRI was acquired from 78 autistic and 81 NT males (12-to-46-years) and fit to the DTI and NODDI diffusion models. TBSS and GBSS were performed to analyze WM and GM microstructure, respectively. General linear models were used to investigate group and age-related group differences. Within the ASD group, relationships between WM and GM microstructure and measures of autistic symptoms were investigated. Results: All dMRI measures were significantly associated with age across WM and GM. Significant group differences were observed across WM and GM. No significant age-by-group interactions were detected. Within the ASD group, positive relationships with WM microstructure were observed with ADOS-2 Calibrated Severity Scores. Conclusion: Using TBSS and GBSS our findings provide new insights into group differences of WM and GM microstructure in autistic males from adolescence into adulthood. Detection of microstructural differences across the lifespan as well as their relationship to the level of autistic symptoms will deepen to our understanding of brain-behavior relationships of ASD and may aid in the improvement of intervention options for autistic adults.