$ exttt{LucidAtlas}$: Learning Uncertainty-Aware, Covariate-Disentangled, Individualized Atlas Representations

2025-02-12

AI Summary

This paper addresses the modeling challenge posed by intertwined spatial heterogeneity, covariate confounding, and population-level uncertainty in high-dimensional medical data. We propose the first spatial atlas learning framework jointly driven by covariate disentanglement and uncertainty awareness. Methodologically, it integrates neural additive models, Bayesian uncertainty quantification, and prior-guided atlas optimization to explicitly disentangle covariate effects; additionally, a marginalization-based interpretability mechanism is introduced to quantitatively attribute covariate-specific modulation of atlas topology. Extensive evaluation on two real-world medical datasets demonstrates the framework’s capabilities in personalized prediction, interpretable covariate attribution, dynamic population trend modeling, and calibrated uncertainty quantification. The implementation code will be made publicly available.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

The goal of this work is to develop principled techniques to extract information from high dimensional data sets with complex dependencies in areas such as medicine that can provide insight into individual as well as population level variation. We develop $ exttt{LucidAtlas}$, an approach that can represent spatially varying information, and can capture the influence of covariates as well as population uncertainty. As a versatile atlas representation, $ exttt{LucidAtlas}$ offers robust capabilities for covariate interpretation, individualized prediction, population trend analysis, and uncertainty estimation, with the flexibility to incorporate prior knowledge. Additionally, we discuss the trustworthiness and potential risks of neural additive models for analyzing dependent covariates and then introduce a marginalization approach to explain the dependence of an individual predictor on the models' response (the atlas). To validate our method, we demonstrate its generalizability on two medical datasets. Our findings underscore the critical role of by-construction interpretable models in advancing scientific discovery. Our code will be publicly available upon acceptance.

Problem

• Extract insights from high-dimensional medical datasets
• Represent spatially varying data with covariate influence
• Ensure interpretable models for scientific discovery

Innovation

• Uncertainty-aware data representation
• Covariate-disentangled atlas modeling
• Individualized prediction and analysis