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

2025-03-14

AI Summary

Alzheimer’s disease (AD) cognitive outcome prediction is hindered by high missingness rates across multimodal data (e.g., neuroimaging, genomics, proteomics) and insufficient biological interpretability of existing models. To address these challenges, we propose a modality-specific collaborative imputation framework integrated with multi-task learning and explainable AI (XAI), establishing the first AD cognitive-domain prediction model supporting multi-output, many-to-many mapping—jointly modeling executive function, language, memory, and visuospatial abilities. Our method combines modality-adaptive imputation, multimodal feature disentanglement, and attribution-based interpretability analysis. Evaluated on 1,205 subjects, it achieves high predictive accuracy and identifies biologically meaningful neuro-transcriptomic biomarker pairs. This advances mechanistic understanding of AD’s complex pathological pathways and significantly enhances clinical interpretability.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Alzheimer's disease, a neurodegenerative disorder, is associated with neural, genetic, and proteomic factors while affecting multiple cognitive and behavioral faculties. Traditional AD prediction largely focuses on univariate disease outcomes, such as disease stages and severity. Multimodal data encode broader disease information than a single modality and may, therefore, improve disease prediction; but they often contain missing values. Recent"deeper"machine learning approaches show promise in improving prediction accuracy, yet the biological relevance of these models needs to be further charted. Integrating missing data analysis, predictive modeling, multimodal data analysis, and explainable AI, we propose OPTIMUS, a predictive, modular, and explainable machine learning framework, to unveil the many-to-many predictive pathways between multimodal input data and multivariate disease outcomes amidst missing values. OPTIMUS first applies modality-specific imputation to uncover data from each modality while optimizing overall prediction accuracy. It then maps multimodal biomarkers to multivariate outcomes using machine-learning and extracts biomarkers respectively predictive of each outcome. Finally, OPTIMUS incorporates XAI to explain the identified multimodal biomarkers. Using data from 346 cognitively normal subjects, 608 persons with mild cognitive impairment, and 251 AD patients, OPTIMUS identifies neural and transcriptomic signatures that jointly but differentially predict multivariate outcomes related to executive function, language, memory, and visuospatial function. Our work demonstrates the potential of building a predictive and biologically explainable machine-learning framework to uncover multimodal biomarkers that capture disease profiles across varying cognitive landscapes. The results improve our understanding of the complex many-to-many pathways in AD.

Problem

• Predicts multivariate Alzheimer's outcomes using multi-modal data
• Handles missing values in multi-modal disease prediction
• Explains biological relevance of predictive biomarkers via XAI

Innovation

• Modality-specific imputation for missing data
• Multimodal biomarker mapping to outcomes
• Explainable AI for biomarker interpretation