Cross Modal Privacy-Preserving Synthesis and Mixture of Experts Ensemble for Robust ASD Prediction

Karthiga M^1*Revathy J²

• ¹Bannari Amman Institute of Technology Department of Computer Science and Engineering, Sathyamangalam, India
• ²Christ the King Engineering College, Coimbatore, India

Abstract Introduction: Autism Spectrum Disorder (ASD) diagnosis remains complex due to limited access to large-scale multimodal datasets and privacy concerns surrounding clinical data. Traditional methods rely heavily on resource-intensive clinical assessments and are constrained by unimodal or non-adaptive learning models. To address these limitations, this study introduces AutismSynthGen, a privacy-preserving framework for synthesizing multimodal ASD data and enhancing prediction accuracy. Methods: The proposed system integrates a Multimodal Autism Data Synthesis Network (MADSN), which employs transformer-based encoders and cross-modal attention within a conditional GAN to generate synthetic data across structural MRI, EEG, behavioural vectors, and severity scores. Differential privacy is enforced via DP-SGD (ε ≤ 1.0). A complementary Adaptive Multimodal Ensemble Learning (AMEL) module, consisting of five heterogeneous experts and a gating network, is trained on both real and synthetic data. Evaluation is conducted on ABIDE, NDAR, and SSC datasets using metrics such as AUC, F1 score, MMD, KS statistic, and BLEU. Results: Synthetic augmentation improved model performance with validation AUC gains of ≥ 0.04. AMEL achieved AUC = 0.98 and F1 = 0.99 on real data, and approached near-perfect internal performance (AUC ≈ 1.00, F1 ≈ 1.00) when synthetic data were included. Distributional metrics (MMD = 0.04; KS = 0.03) and text similarity (BLEU = 0.70) demonstrated high fidelity between real and synthetic samples. Ablation studies confirmed the importance of cross-modal attention and entropy-regularized expert gating. Discussion: AutismSynthGen offers a scalable, privacy-compliant solution for augmenting limited multimodal datasets and enhancing ASD prediction. Future directions include semi-supervised learning, explainable AI for clinical trust, and deployment in federated environments to broaden accessibility while maintaining privacy.