Citation

Sungheetha, Akey and R., Rajesh Sharma (2026) Enhanced Schizophrenia Prediction Using Cross-Modal Attention Mechanisms and EEG Microstate Analysis. In: 4th international conference on Machine Learning and Data Engineering, ICMLDE 2025, 6 November 2025 - 8 November 2025, Dehradun.

Text 26.pdf - Published Version Restricted to Repository staff only Download (404kB)

Abstract

This research presents a novel multi-modal parametric framework integrating cross-modal attention mechanisms with EEG microstate analysis for enhanced schizophrenia prediction and classification. The proposed methodology addresses critical challenges in early detection through parametric optimization of feature extraction coefficients (α = 0.85), attention weight parameters (β = 0.72), and classification threshold variables (γ = 0.68). Our framework incorporates advanced biomarker identification through resting-state EEG microstates with cross-modal fusion parameters (δ = 0.91) and neural network optimization coefficients (ϵ = 0.83). The methodology demonstrates significant improvements in diagnostic accuracy through parametric tuning of connectivity measures (ζ = 0.76), achieving enhanced precision in distinguishing first-episode schizophrenia, ultra-high-risk individuals, and healthy controls. Implementation results show superior performance with optimized sensitivity parameters (η = 0.89) and specificity coefficients (θ = 0.87), establishing a robust foundation for clinical translation and early intervention strategies in mental health diagnostics. The framework achieved 92.8% classification accuracy, outperforming existing methods by 5.6%, with an AUC-ROC of 0.945, demonstrating significant clinical utility for early schizophrenia detection with processing time requirements of 12.7 ± 2.3 seconds per subject, making it suitable for real-time clinical deployment.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Schizophrenia prediction, Multi-modal analysis, EEG microstates
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	03 Sep 2026 07:17
Last Modified:	03 Sep 2026 07:17
URII:	http://shdl.mmu.edu.my/id/eprint/16637

Downloads

Edit (login required)