Abstract
The computation of image similarity is important for a wide range of analyses in neuroimaging, from decoding to meta-analysis. In many cases the images being compared have empty voxels, but the effects of such empty voxels on image similarity metrics are poorly understood. We present a detailed investigation of the influence of different degrees of image thresholding on the outcome of pairwise image comparison. Given a pair of brain maps for which one of the maps is thresholded, we show that an analysis using the intersection of non-zero voxels across images at a threshold of Z = ±1.0 maximizes accuracy for retrieval of a list of maps of the same contrast, and thresholding up to Z = ±2.0 can increase accuracy as compared to comparison using unthresholded maps. Finally, maps can be thresholded up to to Z = ±3.0 (corresponding to 25% of voxels non-empty within a standard brain mask) and still maintain a lower bound of 90% accuracy. Our results suggest that a small degree of thresholding may improve the accuracy of image similarity computations, and that robust meta-analytic image similarity comparisons can be obtained using thresholded images.
Highlights
The computation of similarity between images is an increasingly important component of neuroimaging analyses
In the present work we examine the effects of thresholding on image similarity metrics
Score Distributions Overall, both strategies to handle empty voxels (CCA and single value imputation (SVI)) exhibited decreasing Pearson and Spearman similarity scores with increasing threshold, and this trend was prevalent whether the thresholding included both positive and negative values (Supplementary Video 1), or just positive values (Supplementary Video 2)
Summary
The computation of similarity between images is an increasingly important component of neuroimaging analyses. In the context of reproducibility, statistical brain maps must be compared to evaluate if a new result has successfully replicated a previous one. One challenge in computation of image similarity is the presence of empty (zero-valued) voxels due to thresholding. The clearest example comes from coordinate-based meta-analysis, where voxels outside of regions with activation peaks will have a zero value. At the time of our analysis, for the 774 publicly available maps in the NeuroVault database, 60 (∼7.7%) had fewer than 25% of nonempty voxels observed within an MNI template brain mask. NeuroVault has implemented the ability to compare a single result map to all others in the database.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
