Computer Vision News - March 2021

9 Autism classified by MRI Figure 5: TI values for all the participants with the segmentation data sets. The group averages with error bars are representing 95% confidence intervals (CIs). In the future? This method could be further improved with white matter and functional neurophysiology data as well as behavioral observation and somatic biomarkers. Quantitative measures that can create a distinction with higher detail for ASDs and TDs can be implemented to increase its diagnostic accuracy. An overall diagnostic accuracy of 78.9% for ASD when compared to the clinical interview (gold standard). The novel and straightforward multivariate classification method requires limited expertise (considering machine learning approaches) while still achieving good diagnostic accuracy. It shows promise, adding to earlier research by being able to classify psychiatric disorders - even such a heterogeneous disorder as ASD - using neuroimaging methods. Wrapping up! I hope that the traditional statistical model was useful as a gentle introduction to computational neurobiology and ASD. In the following months, work from computational psychology and neuroscience will be introduced with some of the state-of-the-art machine learning approaches used for classification of ASD and other comorbidities such as ADHD. Until next month, take care!