Algorithm classifies patients with autism spectrum disorder

Wednesday, November 29 | 9:40 a.m.-9:50 a.m. | W3-SSNR11-2 | Room E353BA machine learning-based method for brain MRI exams can classify patients with autism spectrum disorder (ASD), according to this scientific paper.Presenter Mostafa Abdelrahim of the University of Louisville in Louisville, KY, and colleagues developed a two-step, machine-learning framework that extracts morphological markers and then, based on correlation with social responsiveness scale (SRS) behavioral scores, diagnoses subjects as either typical development or ASD. If a patient is deemed to be autistic, the models then determine the severity of the ASD.In testing on the Autism Brain Imaging Data Exchange-II (ABIDE-II) dataset of 521 autistic and 593 control subjects, their method yielded 98.2% sensitivity, 95.6% specificity, and 94% accuracy.“It's worth noting that our approach is not limited to SRS reports and can also be used with other behavioral reports, such as the Autism Diagnostic Observation Schedule (ADOS) reports,” the authors wrote. “Furthermore, the proposed system is not limited to ASD, and it can be utilized for the diagnosis of any brain disorder.”Head over to the Lakeside Center on Wednesday morning to learn more.
Source: AuntMinnie.com Headlines - Category: Radiology Authors: Tags: 2023 AI Preview Source Type: news