Saved in:
Bibliographic Details
Main Authors: Baek, Seunghun, Lee, Jaejin, Sim, Jaeyoon, Jeong, Minjae, Kim, Won Hwa
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.21092
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912729939312640
author Baek, Seunghun
Lee, Jaejin
Sim, Jaeyoon
Jeong, Minjae
Kim, Won Hwa
author_facet Baek, Seunghun
Lee, Jaejin
Sim, Jaeyoon
Jeong, Minjae
Kim, Won Hwa
contents Various neuroimaging studies suffer from small sample size problem which often limit their reliability. Meta-analysis addresses this challenge by aggregating findings from different studies to identify consistent patterns of brain activity. However, traditional approaches based on keyword retrieval or linear mappings often overlook the rich hierarchical structure in the brain. In this work, we propose a novel framework that leverages hyperbolic geometry to bridge the gap between neuroscience literature and brain activation maps. By embedding text from research articles and corresponding brain images into a shared hyperbolic space via the Lorentz model, our method captures both semantic similarity and hierarchical organization inherent in neuroimaging data. In the hyperbolic space, our method performs multi-level neuroimaging meta-analysis (MNM) by 1) aligning brain and text embeddings for semantic correspondence, 2) guiding hierarchy between text and brain activations, and 3) preserving the hierarchical relationships within brain activation patterns. Experimental results demonstrate that our model outperforms baselines, offering a robust and interpretable paradigm of multi-level neuroimaging meta-analysis via hyperbolic brain-text representation.
format Preprint
id arxiv_https___arxiv_org_abs_2511_21092
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MNM : Multi-level Neuroimaging Meta-analysis with Hyperbolic Brain-Text Representations
Baek, Seunghun
Lee, Jaejin
Sim, Jaeyoon
Jeong, Minjae
Kim, Won Hwa
Machine Learning
Artificial Intelligence
Various neuroimaging studies suffer from small sample size problem which often limit their reliability. Meta-analysis addresses this challenge by aggregating findings from different studies to identify consistent patterns of brain activity. However, traditional approaches based on keyword retrieval or linear mappings often overlook the rich hierarchical structure in the brain. In this work, we propose a novel framework that leverages hyperbolic geometry to bridge the gap between neuroscience literature and brain activation maps. By embedding text from research articles and corresponding brain images into a shared hyperbolic space via the Lorentz model, our method captures both semantic similarity and hierarchical organization inherent in neuroimaging data. In the hyperbolic space, our method performs multi-level neuroimaging meta-analysis (MNM) by 1) aligning brain and text embeddings for semantic correspondence, 2) guiding hierarchy between text and brain activations, and 3) preserving the hierarchical relationships within brain activation patterns. Experimental results demonstrate that our model outperforms baselines, offering a robust and interpretable paradigm of multi-level neuroimaging meta-analysis via hyperbolic brain-text representation.
title MNM : Multi-level Neuroimaging Meta-analysis with Hyperbolic Brain-Text Representations
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2511.21092