Meta-Analytic Techniques in Neuroimaging Studies

Meta-analytic techniques have become indispensable for synthesising results across the burgeoning literature of human brain imaging. Two principal approaches prevail: image-based meta-analysis, which aggregates statistical maps from individual studies, and coordinate-based meta-analysis (CBMA), which pools reported peak coordinates when full images are unavailable. Among CBMA methods, Activation Likelihood Estimation (ALE) and seed-based d Mapping with Permutation of Subject Images (SDM-PSI) have gained widespread adoption. These algorithms model spatial uncertainty around activation foci, estimate convergence across experiments and apply statistical thresholds to distinguish robust effects from chance clusters. Recent efforts have addressed key challenges, including automated pipelines for data curation, improved thresholding strategies to control false positives, and integration of multimodal data such as transcriptomic profiles. Advances in reproducibility and open-source toolboxes now allow researchers to conduct end-to-end meta-analyses more efficiently, while sensitivity analyses and Bayesian frameworks help assess the stability of findings. Collectively, these developments have enhanced the capacity to map consistent brain-behaviour relationships, identify disease-related patterns and guide hypothesis generation for future neuroimaging research.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Research from all publishers

Innovations in thresholding for ALE have introduced a Bayes factor-based procedure that translates conventional frequentist cut-offs into a minimum Bayes factor (mBF) framework. This approach permits graded inference about activation clusters, offering researchers the flexibility to interpret weaker signals without sacrificing statistical rigour.

A new MATLAB toolbox for CBMA automates data preparation and post-hoc analyses, reducing manual errors and accelerating reproducible workflows. By enabling standardised kernel creation, coordinate conversion and cluster-level statistics, the toolbox streamlines every stage from raw coordinate inputs to final meta-analytic maps.

Seed-based d Mapping with Permutation of Subject Images has been deployed to compare structural magnetic resonance imaging profiles in late-life depression and mild cognitive impairment, integrating voxel-based morphometry outcomes with gene-expression data. This work exemplifies how meta-analytic pipelines can extend beyond mere convergence mapping to reveal neurobiological and molecular substrates underlying clinical syndromes.

Meta-Analytic Techniques in Neuroimaging Studies publication trend

The graph below shows the total number of articles in meta-analytic techniques in neuroimaging studies across all publications each year (not limited to Nature Index journals).

Technical terms

Coordinate-Based Meta-Analysis (CBMA): A method that aggregates reported peak activation coordinates from multiple neuroimaging studies to identify spatial convergence of brain activity or structural differences.

Activation Likelihood Estimation (ALE): A statistical algorithm within CBMA that models each reported coordinate as a probability distribution and computes the likelihood of activation across studies.

Seed-based d Mapping with Permutation of Subject Images (SDM-PSI): A voxel-based meta-analysis technique that combines peak coordinates and effect sizes, using permutation tests to assess statistical significance.

Voxel-Based Morphometry (VBM): A whole-brain analysis technique that measures local concentrations of grey matter and white matter, often used as input for structural meta-analyses.

Bayes Factor (mBF) Thresholding: A Bayesian approach to determining statistical thresholds, expressing evidence for or against hypotheses in terms of Bayes factors rather than p-values.

References

1. A Minimum Bayes Factor Based Threshold for Activation Likelihood Estimation. Neuroinformatics (2023).
2. CBMAT: a MATLAB toolbox for data preparation and post hoc analyses in neuroimaging meta-analyses. Behavior Research Methods (2023).
3. A comparative meta-analysis of structural magnetic resonance imaging studies and gene expression profiles revealing the similarities and differences between late life depression and mild cognitive impairment. Psychological Medicine (2024).