COSTA MESA, Calif., April 28, 2017 /PRNewswire/ -- The diagnosis of Autism Spectrum Disorder (ASD) relies on history and behavioral observation, lacking reliable biomarkers. Researchers from Amen Clinics and the University of Southern California performed what is believed to be the largest ever analysis of brain SPECT (single photon emission computed tomography) scans, a nuclear medicine study that evaluates blood flow and activity patterns, on 928 persons with ASD obtained 9 different sites to investigate whether these scans distinguish ASD from healthy controls. The age range of patients were from 13-67 years.

Using sophisticated machine learning algorithms, high levels of separation were obtained. The areas the most predicted ASD were found in the cerebellum, anterior cingulate gyrus, amygdala, frontal and temporal lobes.

Lead author Daniel Amen, MD, child psychiatrist and founder of Amen Clinics said, "Currently, the diagnosis of ASD includes a clinical history, mental status examination and structured screening tools, leaving clinicians in the dark as to the underlying physiology. At Amen Clinics, we frequently see increased activity in the anterior cingulate, leading to obsessive behavior, and decreases in the temporal lobes and cerebellum, which are often associated with learning issues. Having SPECT scans on ASD patients has helped us better target treatment."
This is the first brain SPECT imaging study demonstrating the use of machine learning methods to predict ASD from a HC. These results add to the growing body of literature validating the use of machine learning approaches with functional neuroimaging data to improve prediction and classification of individuals with psychiatric disorders like autism. Given the heterogeneity of ASD, this approach has important implications in the clinical setting in both the diagnosis, intervention and monitoring of treatment outcomes.


SOURCE Amen Clinics, Inc.

Molecular Imaging Homepage