Abstract:Background: The integration of the information encoded in multiparametric MRI images can enhance the performance of machine-learning classifiers. In this study, we investigate whether the combination of structural and functional MRI might improve the performances of a deep learning (DL) model trained to discriminate subjects with Autism Spectrum Disorders (ASD) with respect to typically developing controls (TD). Material and methods: We analyzed both structural and functional MRI brain scans publicly available within the ABIDE I and II data collections. We considered 1383 male subjects with age between 5 and 40 years, including 680 subjects with ASD and 703 TD from 35 different acquisition sites. We extracted morphometric and functional brain features from MRI scans with the Freesurfer and the CPAC analysis packages, respectively. Then, due to the multisite nature of the dataset, we implemented a data harmonization protocol. The ASD vs. TD classification was carried out with a multiple-input DL model, consisting in a neural network which generates a fixed-length feature representation of the data of each modality (FR-NN), and a Dense Neural Network for classification (C-NN). Specifically, we implemented a joint fusion approach to multiple source data integration. The main advantage of the latter is that the loss is propagated back to the FR-NN during the training, thus creating informative feature representations for each data modality. Then, a C-NN, with a number of layers and neurons per layer to be optimized during the model training, performs the ASD-TD discrimination. The performance was evaluated by computing the Area under the Receiver Operating Characteristic curve within a nested 10-fold cross-validation. The brain features that drive the DL classification were identified by the SHAP explainability framework. Results: The AUC values of 0.66±0.05 and of 0.76±0.04 were obtained in the ASD vs. TD discrimination when only structural or functional features are considered, respectively. The joint fusion approach led to an AUC of 0.78±0.04. The set of structural and functional connectivity features identified as the most important for the two-class discrimination supports the idea that brain changes tend to occur in individuals with ASD in regions belonging to the Default Mode Network and to the Social Brain. Conclusions: Our results demonstrate that the multimodal joint fusion approach outperforms the classification results obtained with data acquired by a single MRI modality as it efficiently exploits the complementarity of structural and functional brain information.

Research on multi‐center assisted diagnosis of ASD based on multimodal feature fusion

Multi-task Diagnosis for Autism Spectrum Disorders Using Multi-Modality Features: A Multi-Center Study.

Multi-Class ASD Classification Based on Functional Connectivity and Functional Correlation Tensor via Multi-Source Domain Adaptation and Multi-View Sparse Representation

A multi-view convolutional neural network method combining attention mechanism for diagnosing autism spectrum disorder

Multi-head Self-Attention Mechanism-Based Global Feature Learning Model for ASD Diagnosis

Fusion of Fmri Multisource Data Using a Two-stream Separable 3D Network for the Diagnosis of Autism Spectrum Disorder

Attention-Like Multimodality Fusion With Data Augmentation for Diagnosis of Mental Disorders Using MRI

Multi-kernel Learning Fusion Algorithm Based on RNN and GRU for ASD Diagnosis and Pathogenic Brain Region Extraction

SSMDA : Semi-supervised multi-source domain adaptive autism prediction model using neuroimaging

A Multimodal Approach for Identifying Autism Spectrum Disorders in Children

MAFT-SO: A novel multi-atlas fusion template based on spatial overlap for ASD diagnosis

AIMAFE: Autism Spectrum Disorder Identification with Multi-Atlas Deep Feature Representation and Ensemble Learning

Multi-scale Convolutional Network for fMRI-Based Diagnosis of Autism Spectrum Disorder

Self-weighted Adaptive Structure Learning for ASD Diagnosis Via Multi-Template Multi-Center Representation

Multi-site Diagnostic Classification of Autism Spectrum Disorder Using Adversarial Deep Learning on Resting-State Fmri.

Deep learning based joint fusion approach to exploit anatomical and functional brain information in autism spectrum disorders

Low-rank Domain Adaptive Method with Inter-class difference Constraint for Multi-site Autism Spectrum Disorder Identification

Deep Multimodal Learning for the Diagnosis of Autism Spectrum Disorder

Combining Autoencoder and Category-Based Low-Rank Domain Adaptation Method for Multi-Site ASD Identification

Interpretable Feature Learning Using Multi-output Takagi-Sugeno-Kang Fuzzy System for Multi-center ASD Diagnosis.

Enhancing Autism Spectrum Disorder identification in multi-site MRI imaging: A multi-head cross-attention and multi-context approach for addressing variability in un-harmonized data