Advanced meditation brain states decoded using multi atlas spectral method
ACAM-J-MARS: Multi-Atlas Riemannian Spectral Representations for Decoding Advanced Meditation States from fMRI
Machine Learning
Summary
It is hard to tell different deep meditation states apart using brain scans because effects are spread across the brain and vary a lot between people. The authors designed a new method called ACAM-J-MARS that looks at brain connections using several brain maps, treats brain data in a way that respects its complex geometry, and includes information about slow brain activity patterns. This approach improves accuracy in distinguishing early and late meditation states and helps better understand brain patterns during meditation. It could also help build tools that give real-time feedback on brain states during meditation.
What this means in practice
- •For neurofeedback developers: Create systems that provide real-time feedback on advanced meditation states using improved brain connectivity representations from fMRI data.
- •For brain-computer interface engineers: Build real-time psychophysiological interfaces that decode subtle meditation states using geometric and spectral brain data features.
Authors
Jian Wang, Winson F. Z. Yang, Wugeng Zheng, Matthew D. Sacchet
Abstract
Inferring jhana-type advanced concentration absorption meditation (ACAM-J) states from fMRI data is challenging due to the spatially distributed nature of meditation-related effects, substantial inter-individual variability that can exceed within-subject state differences, and limited sample sizes. Conventional machine-learning approaches often formulate this problem as unimodal, Euclidean, atlas-based classification, thereby neglecting the non-Euclidean geometry of functional connectivity matrices and underrepresenting temporal information in functional magnetic resonance imaging (fMRI); these limitations can lead to suboptimal classification performance. We propose ACAM-J-MARS (\textbf{M}ulti-\textbf{A}tlas \textbf{R}iemannian \textbf{S}pectral representation), which computes functional connectivity using multiple atlases, represents symmetric positive definite (SPD) connectivity matrices on a Riemannian manifold, projects them into a log-Euclidean tangent space, and augments connectivity-derived features with low-frequency BOLD spectral energy to better resolve subtle state differences under heterogeneous variability. Evaluated on 204 segments from 22 practitioners, ACAM-J-MARS improves performance across multiple backbone models, achieving a balanced accuracy of 0.757 for early versus late ACAM-J classification and 0.545 for three-class classification, corresponding to a 10.8-point absolute improvement over the strongest baseline. These findings suggest that the proposed representation improves the decoding of advanced meditative brain states and may enable downstream applications such as closed-loop neurofeedback and real-time psychophysiological brain-computer interfaces.