High-dimensional multimodal uncertainty estimation by manifold alignment:Application to 3D right ventricular strain computations

2025-01-21

AI Summary

Quantifying local uncertainty in high-dimensional, multi-source physiological descriptors—such as three-dimensional right ventricular (RV) strain—is challenging due to inter-method discrepancies in definition and computation. Method: We propose the first uncertainty propagation framework integrating manifold alignment with variational latent distribution modeling. Unlike conventional approaches that model only epistemic uncertainty, our method enables uncertainty propagation and nonlinear inverse reconstruction within a cross-modal, high-dimensional latent space, supporting fusion analysis of heterogeneous strain tensors. Results: Evaluated on 100 RV overload patients, our framework achieves, for the first time, pointwise quantification of local uncertainty in both direction and magnitude of endocardial strain. This provides clinicians with interpretable, verifiable, and trustworthy AI-driven decision support to address inconsistencies arising from divergent strain definitions.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Uncertainty Representations
• Intelligent Robots: State Estimation

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

Confidence in the results is a key ingredient to improve the adoption of machine learning methods by clinicians. Uncertainties on the results have been considered in the literature, but mostly those originating from the learning and processing methods. Uncertainty on the data is hardly challenged, as a single sample is often considered representative enough of each subject included in the analysis. In this paper, we propose a representation learning strategy to estimate local uncertainties on a physiological descriptor (here, myocardial deformation) previously obtained from medical images by different definitions or computations. We first use manifold alignment to match the latent representations associated to different high-dimensional input descriptors. Then, we formulate plausible distributions of latent uncertainties, and finally exploit them to reconstruct uncertainties on the input high-dimensional descriptors. We demonstrate its relevance for the quantification of myocardial deformation (strain) from 3D echocardiographic image sequences of the right ventricle, for which a lack of consensus exists in its definition and which directional component to use. We used a database of 100 control subjects with right ventricle overload, for which different types of strain are available at each point of the right ventricle endocardial surface mesh. Our approach quantifies local uncertainties on myocardial deformation from different descriptors defining this physiological concept. Such uncertainties cannot be directly estimated by local statistics on such descriptors, potentially of heterogeneous types. Beyond this controlled illustrative application, our methodology has the potential to be generalized to many other population analyses considering heterogeneous high-dimensional descriptors.

Problem

• Uncertainty Estimation
• Cardiac Deformation
• Machine Learning Trust

Innovation

• Uncertainty Estimation
• Cardiac Muscle Deformation
• High-Dimensional Data Analysis