European Radiology: Deep Learning Makes Autism Diagnosis Possible Using Conventional MRI!

 Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder characterized by impairments in socialization and communication, as well as repetitive and restrictive behaviors

 Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder characterized by impairments in socialization and communication, as well as repetitive and restrictive behaviors

Conventional MRI (cMRI) and diffusion-weighted imaging (DWI) are widely used non-invasive methods

In recent years, with the continuous progress of deep learning (DL) algorithms, some studies have found that MRI-based DL algorithms can be used for the identification of ASD and TD, and even for the prediction of ASD

Recently, a study published in the journal European Radiology constructed a DL-based ASD diagnostic model using cMRI (including axial T1, T2, FLAIR , and sagittal T1/T2 sequences) and brain ADC images from an age-matched cohort .

A total of 151 children with ASD and 151 age-matched typically developmental (TD) control children were included in this study

The highest AUCs (0.

Figure Grad-CAM-assisted ASD image recognition .

Figure Grad-CAM-assisted ASD image recognition .

The results of this study demonstrate that the identification of children with ASD from children with TD can be achieved based on commonly used MRI sequences .

Original source :

Xiang Guo , Jiehuan Wang , Xiaoqiang Wang , et al .

Xiang Guo Jiehuan Wang Xiaoqiang Wang ,et al Diagnosing autism spectrum disorder in children using conventional MRI and apparent diffusion coefficient based deep learning algorithms Diagnosing autism spectrum disorder in children using conventional MRI and apparent diffusion coefficient based deep learning algorithms 10.
1007/s00330-021-08239 -4 10.
1007/s00330-021-08239-4

Leave a message here