De-Individualizing fMRI Signals via Mahalanobis Whitening and Bures Geometry

2025-11-10

AI Summary

Functional MRI (fMRI) functional connectivity (FC) in preclinical Alzheimer’s disease (AD) diagnosis suffers from substantial inter-subject variability and poor cross-subject comparability due to individual differences. Method: This paper proposes a Bures-geometry-based individualization-removal framework. Its core innovation reinterprets Mahalanobis whitening as a two-stage Bures distance minimization process and incorporates quantum fidelity—a geometrically grounded, physically interpretable metric—for data standardization. The method integrates whitening preprocessing with Bures-metric-driven dimensionality reduction to enhance FC matrix consistency across subjects. Contribution/Results: Experiments demonstrate that the framework effectively disentangles stimulus-specific neural responses from subject-specific noise, significantly improving the robustness and reproducibility of early AD biomarker detection. By providing a geometrically principled, biophysically informed approach to FC standardization, it establishes a novel paradigm for precision neuroimaging diagnosis based on functional connectivity.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Computer Vision: Other Foundations of Computer Vision
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Web data integration and cleaning
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Functional connectivity has been widely investigated to understand brain disease in clinical studies and imaging-based neuroscience, and analyzing changes in functional connectivity has proven to be valuable for understanding and computationally evaluating the effects on brain function caused by diseases or experimental stimuli. By using Mahalanobis data whitening prior to the use of dimensionality reduction algorithms, we are able to distill meaningful information from fMRI signals about subjects and the experimental stimuli used to prompt them. Furthermore, we offer an interpretation of Mahalanobis whitening as a two-stage de-individualization of data which is motivated by similarity as captured by the Bures distance, which is connected to quantum mechanics. These methods have potential to aid discoveries about the mechanisms that link brain function with cognition and behavior and may improve the accuracy and consistency of Alzheimer's diagnosis, especially in the preclinical stage of disease progression.

Problem

• De-individualizing fMRI signals using Mahalanobis whitening and Bures geometry
• Distilling meaningful information from fMRI data about subjects and stimuli
• Improving Alzheimer's diagnosis accuracy through functional connectivity analysis

Innovation

• Mahalanobis whitening for fMRI signal processing
• Bures geometry for data de-individualization
• Dimensionality reduction for functional connectivity analysis