Exploring sMRI Biomarkers for Diagnosis of Autism Spectrum Disorders Based on Multi Class Activation Mapping Models

Abstract

Due to the complexity of the etiology of autism spectrum disorders, the existing autism diagnosis method is still based on scales. With the continuous development of artificial intelligence, imageaided diagnosis of brain diseases has been widely studied and concerned. However, many doctors and researchers still doubt the diagnosis basis of the neural network and think that the neural network belongs to a limited interpretable black-box function approximator. They are not sure whether the neural network has learned some interpretive image features like humans. In order to solve this problem, three new models (2D CAM, 3D CAM and 3D Grad-CAM) are proposed for structural Magnetic Resonance Imaging (sMRI) data. The Regions Of Interest (ROI) of subcortical tissues among models and between groups are analyzed based on the heat maps of the three models. The experimental results show that these models mainly distinguish the autism group and the control group according to the voxel value of these ROIs. There are significant differences in mean voxel value and standard deviation of voxel value between the autism group and the control group, such as in the left amygdala, optic chiasm and right hippocampus. According to medical references, these ROIs are closely related to people's speech, cognition and behavior. This can partly explain why autistic patients have unusual symptoms such as speech communication disorder, stereotyped repetitive behavior and so on. The proposed visualization models can provide a good bridge for doctors to understand the brain features learned by the neural network. The research method of this paper may provide a new way for doctors and researchers to find the diagnostic biomarkers of autism, which can greatly speed up the process of modern medical diagnosis and treatment strategies, and liberate doctors from the traditional trial and error.

Automated diagnosis of autism: State of the art

“…In the study of Yang 2021 [337], three new models (2D CAM, 3D CAM, and 3D Grad-CAM) were proposed which help with the interpretability of the classification task on S-MRI data.…”

Differentiating Pressure Ulcer Risk Levels through Interpretable Classification Models Based on Readily Measurable Indicators

“…Consequently, various XAI tools have been employed in the medical domain for diagnostics. For example, Gradient-weighted Class Activation Mapping (Grad-CAM), which has been applied in medical areas such as autism spectrum disorder [31] and brain tumor detection [32], uses gradient information from a CNN-based architecture to create a map highlighting the important regions in an image based on the classification. Local Interpretable Model-agnostic Explanations (LIME) is another notable XAI method that provides a local estimation for interpreting individual predictions, illustrating the impact of each feature on the model's outcome, and has been applied to conditions such as Alzheimer's and Parkinson's diseases [33,34].…”