Identifying Neural Signatures from fMRI using Hybrid Principal Components Regression

2025-09-08

AI Summary

Conventional LASSO-PCR for fMRI decoding assumes equal importance of all principal components (PCs), limiting adaptive selection of task-relevant components. Method: We propose Joint Sparse Ranking LASSO (JSRL), which, following PCA dimensionality reduction, integrates component-level regularization—ranked by explained variance—with voxel-level sparsity constraints to achieve balanced information fusion. Results: Evaluated on multi-task fMRI data involving risk decision-making, reward processing, and emotion regulation, JSRL significantly improves psychological state decoding: cross-validated R² increases by up to 51.7%, and AUC improves by 7.3%. Learned weight maps align with established functional neuroanatomy, confirming neurobiological interpretability. Crucially, JSRL is the first method to jointly optimize PC selection and voxel-wise contribution via coordinated sparsity, establishing a novel paradigm for high-dimensional neuroimaging decoding.

Technology Category

• Machine Learning: Learning Preferences or Rankings
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Computer Vision: Learning & Optimization for CV

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Recent advances in neuroimaging analysis have enabled accurate decoding of mental state from brain activation patterns during functional magnetic resonance imaging scans. A commonly applied tool for this purpose is principal components regression regularized with the least absolute shrinkage and selection operator (LASSO PCR), a type of multi-voxel pattern analysis (MVPA). This model presumes that all components are equally likely to harbor relevant information, when in fact the task-related signal may be concentrated in specific components. In such cases, the model will fail to select the optimal set of principal components that maximizes the total signal relevant to the cognitive process under study. Here, we present modifications to LASSO PCR that allow for a regularization penalty tied directly to the index of the principal component, reflecting a prior belief that task-relevant signal is more likely to be concentrated in components explaining greater variance. Additionally, we propose a novel hybrid method, Joint Sparsity-Ranked LASSO (JSRL), which integrates component-level and voxel-level activity under an information parity framework and imposes ranked sparsity to guide component selection. We apply the models to brain activation during risk taking, monetary incentive, and emotion regulation tasks. Results demonstrate that incorporating sparsity ranking into LASSO PCR produces models with enhanced classification performance, with JSRL achieving up to 51.7% improvement in cross-validated deviance $R^2$ and 7.3% improvement in cross-validated AUC. Furthermore, sparsity-ranked models perform as well as or better than standard LASSO PCR approaches across all classification tasks and allocate predictive weight to brain regions consistent with their established functional roles, offering a robust alternative for MVPA.

Problem

• Improving principal component selection in fMRI analysis
• Enhancing neural signature identification accuracy
• Addressing suboptimal component weighting in LASSO PCR

Innovation

• Hybrid JSRL method integrates component and voxel level sparsity
• Ranked sparsity regularization based on principal component indices
• Information parity framework guides optimal component selection