MADE-for-ASD: A Multi-Atlas Deep Ensemble Network for Diagnosing Autism Spectrum Disorder

TL;DR

This paper targets efficient, objective ASD diagnosis from resting-state fMRI by introducing MADE-for-ASD, a multi-atlas deep ensemble framework that fuses functional connectivity features from three brain atlases with demographic information. The model integrates a stacked sparse denoising autoencoder for unsupervised pretraining with a transfer-learned MLP and a weighted soft-voting ensemble, achieving 75.20% accuracy on ABIDE I and 96.40% on the NYU subset, outperforming prior ABIDE I benchmarks. AF-score guided feature selection and a data-driven subset quality selector (trained on NYU) bolster robustness, while visualizations highlight key ROIs such as the precuneus and anterior cingulate regions as informative for ASD. The approach promises scalable, cost-effective screening and provides neurobiological interpretability cues, with future work addressing interpretability and demographic biases.

Abstract

In response to the global need for efficient early diagnosis of Autism Spectrum Disorder (ASD), this paper bridges the gap between traditional, time-consuming diagnostic methods and potential automated solutions. We propose a multi-atlas deep ensemble network, MADE-for-ASD, that integrates multiple atlases of the brain's functional magnetic resonance imaging (fMRI) data through a weighted deep ensemble network. Our approach integrates demographic information into the prediction workflow, which enhances ASD diagnosis performance and offers a more holistic perspective on patient profiling. We experiment with the well-known publicly available ABIDE (Autism Brain Imaging Data Exchange) I dataset, consisting of resting state fMRI data from 17 different laboratories around the globe. Our proposed system achieves 75.20% accuracy on the entire dataset and 96.40% on a specific subset $-$ both surpassing reported ASD diagnosis accuracy in ABIDE I fMRI studies. Specifically, our model improves by 4.4 percentage points over prior works on the same amount of data. The model exhibits a sensitivity of 82.90% and a specificity of 69.70% on the entire dataset, and 91.00% and 99.50%, respectively, on the specific subset. We leverage the F-score to pinpoint the top 10 ROI in ASD diagnosis, such as precuneus and anterior cingulate/ventromedial. The proposed system can potentially pave the way for more cost-effective, efficient and scalable strategies in ASD diagnosis. Codes and evaluations are publicly available at https://github.com/hasan-rakibul/MADE-for-ASD.

Figures (7)

• Figure 1: Overall framework of our ASD/TC classification workflow. We first calculate functional connectivity matrices on three brain atlases, followed by feature selection using F-score. To diagnose ASD, we use deep learning with a stacked sparse denoising autoencoder (SSDAE) and multi-layer perceptron (MLP), followed by a weighted ensemble. In the case of extracting high-quality data, we train the classifier with the NYU subset and predict on the whole dataset.
• Figure 2: Overall architecture and training workflow of the deep networks of the MADE-for-ASD model. Knowledge from stacked sparse denoising autoencoders is transferred to a multi-layer perceptron for ASD diagnosis. Here, $X$ refers to the input data, $Y = \{Y_1, Y_2\}$ are ASD and TC classes, and $W$ refers to the weight parameter.
• Figure 3: The multi-atlas weighted ensemble voting strategy.
• Figure 4: Top 10 most significant regions of interest towards ASD diagnosis using fMRI CC200 atlas.
• Figure A.5: ASD classification accuracy with different feature selection ranges in our F-score-based feature selection. The presented accuracy refers to the average accuracy over all three parcellations.