Computer Aided Autism Diagnosis Using Diffusion Tensor Imaging

Abstract

Autism Spectrum Disorder (ASD), commonly known as autism, is a lifelong developmental disorder associated with a broad range of symptoms including difficulties in social interaction, communication skills, and restricted and repetitive behaviors. In autism, numerous studies suggest abnormal development of neural networks that manifest itself as abnormalities of brain shape, functionality, and/ or connectivity. The aim of this work is to present our automated computer aided diagnostic (CAD) system for accurate identification of autism based on the connectivity of the white matter (WM) tracts. To achieve this goal, two levels of analysis are provided for local and global scores using diffusion tensor imaging (DTI) data. A local analysis using Johns Hopkins WM areas' atlas is exploited for DTI atlas-based segmentation. Furthermore, WM integrity is examined by extracting the most notable features representing WM connectivity from DTI. Interactions of WM features between different areas in the brain, demonstrating correlations between WM areas were used, and feature selection among those associations were made. Finally, a LOSO classifier is employed to yield a final per-subject decision. The proposed system was tested on a large dataset of 263 subjects from NDAR database with their Autism Diagnostic Observation Schedule (ADOS) scores and diagnosis (141 typically developed: 66 males, and 75 females, and 122 autistics: 66 males, and 56 females), with ages ranging from 96 to 215 months, achieving an overall accuracy of 73%. In addition to this achieved global accuracy, diagnostically-important brain areas were identified, allowing for a better understanding of ASD-related brain abnormalities, which is considered as an essential step towards developing early personalized treatment plans for children with autism.

The Role of Diffusion Tensor MR Imaging (DTI) of the Brain in Diagnosing Autism Spectrum Disorder: Promising Results

“…Reduced FA, increased RD, and decreased AD of the tapetum has been reported in ASD. Abnormalities in the corticospinal tract, corona radiata, external capsule, cingulum cingulate cyrus, cingulum hippocampus, and superior fronto-occipital fasciculus were noted in previous studies [13,23,40,42,44,49,[52][53][54][55][56]. We stress that our findings are for brain regions' interactions with others, following the idea of disrupted connectivity introduced by Vasa et al, and work normally when done in functional MRI experiments.…”

“…While most of those works relied on sMRI and/or fMRI, the focus of our paper is using DTI. DTI micro-architectural features were incorporated in another large recent study on 263 NDAR subjects for the diagnosis of autism, achieving accuracy of up to 73% [23]. Up to now, most of the published work regarding autism classification used ABIDE-I, and very few studies used newer ABIDE-II data [21,[24][25][26].…”

Autism Spectrum Disorder Diagnosis Assistance using Machine Learning

“…Para isso eles mesmos selecionaram crianças, cujo diagnóstico do TEA já tivesse sido realizado, e aplicaram algoritmos para detalhar as características faciais de cada indivíduo criando, assim, uma BD. Após esse processo, os dados foram submetidos a três algoritmos de AM, sendo eles: SVM, RF e Neural Networks Multilayer Perceptron (MLP), descrevendo dois cenários distintos, onde o primeiro cenário foram empregadas [32] Genéticos SVM, Naive Bayes (NB), LDA, KNN Liu et al (2020) [33] Imagens ressônancia magnética SVM Elnakieb et al (2020) [34] Imagens ressônancia magnética SVM, DT, KNN, Neural Network (NN) Huang et al (2019) [35] Imagens ressônancia magnética SVM ElNakieb et al (2019) [36] Imagens ressônancia magnética SVM, KNN, DT, NN, Deep Neural Network (DNN) Mostafa, Tang and Wu (2019) [37] Imagens ressônancia magnética LDA, Logistic Regression (LR), SVM, KNN, NN Haputhanthri et al (2019) [38] Voz por EGG SVM, LR, RF, NB Wu et al (2021) [39] Imagens Vídeo SVM Sidhu (2019) [40] Imagens ressônancia magnética Principal Component Analysis (PCA), Independent Component Analysis (ICA), SVM Hasan, Jailani and Tahir (2018) [41] Imagens 3D movimentos KNN, SVM, NN Vijayalakshmi et al (2020) [42] Comportamentais demográficos NB, RF, LR Akter et al (2019) [43] Comportamentais biológicos SVM, DT, LR Roopa and Prasad (2019) [44] Imagens ressônancia magnética SVM, RF, DNN Huang, Liu and Tan (2020) [45] Imagens ressônancia magnética SVM Aslam et al (2021) [46] Imagens [48].…”

Autism Detection of MRI Brain Images Using Hybrid Deep CNN With DM-Resnet Classifier

“…In pathophysiology ASD, disrupted connectivity theory is considered as a key feature. From the brain image of ASD individuals, the reduced information transfer is perceived as a consequence of long range under connectivity and local over connectivity [10]. In between various anatomical regions, the pattern is detected with a functional connectivity measure.…”