This research introduces OPTIMUS, an innovative approach that leverages multiple data types simultaneously to predict various Alzheimer's disease outcomes, even when data is incomplete.

• Combines neural, genetic, and proteomic data to improve prediction accuracy
• Addresses the critical challenge of missing values in medical datasets
• Predicts multiple disease outcomes rather than just single variables
• Demonstrates how deep learning can effectively integrate complex biomedical data

This advancement matters for healthcare by potentially enabling earlier, more comprehensive Alzheimer's diagnosis and progression tracking, supporting more personalized treatment approaches and clinical decision-making.

OPTIMUS: Predicting Multivariate Outcomes in Alzheimer's Disease Using Multi-modal Data amidst Missing Values