Quantile Function-Based Models for Neuroimaging Classification Using Wasserstein Regression

2025-08-29

AI Summary

Conventional neuroimaging approaches for mild traumatic brain injury (mTBI) detection rely on summary statistics, discarding critical distributional information—such as variance, skewness, and kurtosis—of neural signals. Method: We propose the first classification framework treating probability density functions (PDFs) as functional response variables, integrating magnetoencephalography (MEG) and magnetic resonance imaging (MRI) data to fully preserve epoch-level signal distributions. Innovatively, we represent distributions via quantile functions and develop a Wasserstein–Fréchet regression model that jointly incorporates age and sex covariates, estimating an L1-minimal-norm solution. Classification is achieved by quantifying inter-group distributional divergence using the Wasserstein distance between healthy controls and mTBI patients. Contribution/Results: Evaluated on the unseen test set of the Innovision IP dataset, our method achieves 98% classification accuracy—substantially improving diagnostic sensitivity and precision for mTBI over existing approaches.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Web Mining and Content Analysis: Content-based information diffusion
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

We propose a novel quantile function-based approach for neuroimaging classification using Wasserstein-Fréchet regression, specifically applied to the detection of mild traumatic brain injury (mTBI) based on the MEG and MRI data. Conventional neuroimaging classification methods for mTBI detection typically extract summary statistics from brain signals across the different epochs, which may result in the loss of important distributional information, such as variance, skewness, kurtosis, etc. Our approach treats complete probability density functions of epoch space results as functional response variables within a Wasserstein-Fréchet regression framework, thereby preserving the full distributional characteristics of epoch results from $L_{1}$ minimum norm solutions. The global Wasserstein-Fréchet regression model incorporating covariates (age and gender) allows us to directly compare the distributional patterns between healthy control subjects and mTBI patients. The classification procedure computes Wasserstein distances between estimated quantile functions from control and patient groups, respectively. These distances are then used as the basis for diagnostic decisions. This framework offers a statistically principled approach to improving diagnostic accuracy in mTBI detection. In practical applications, the test accuracy on unseen data from Innovision IP's dataset achieves up to 98%.

Problem

• Classifying mild traumatic brain injury using MEG and MRI data
• Preserving full distributional information in neuroimaging analysis
• Improving diagnostic accuracy with Wasserstein-Fréchet regression

Innovation

• Quantile function-based approach for neuroimaging classification
• Wasserstein-Fréchet regression with functional response variables
• Classification using Wasserstein distances between distribution patterns