BrainDancer Dynamic fMRI Denoising Enhances Biological Contrasts and State-Dependent Effects
Presented During: Poster Session 3
Friday, June 27, 2025: 01:45 PM - 03:45 PM
Friday, June 27, 2025: 01:45 PM - 03:45 PM
Presented During: Poster Session 4
Saturday, June 28, 2025: 01:45 PM - 03:45 PM
Saturday, June 28, 2025: 01:45 PM - 03:45 PM
Poster No:
1357
Submission Type:
Abstract Submission
Authors:
Helmut Strey1, Zeming Kuang2, Lilianne Mujica-Parodi1
Institutions:
1State University of New York at Stony Brook, Stony Brook, NY, 2Stony Brook University, Stony Brook, MA
First Author:
Co-Author(s):
Introduction:
Recent neuroimaging protocols have prioritized test-retest reliability, potentially at the expense of capturing meaningful biological variance. While subject-level reliability remains important, there is a critical need to preserve and enhance biologically relevant trait and state variations (Finn, 2021). We developed BrainDancer, a dynamic phantom-based approach for data-driven denoising that amplifies biological contrasts while maintaining data integrity (Kumar, 2021).
Methods:
We evaluated BrainDancer's effectiveness using two 7-Tesla fMRI datasets. Dataset 1 (n=32, ages 18-44) included a dot category match task. The participants were asked to find a matching dot pattern to one of two known categories. One dot was presented at a time, and the faster the participant made the correct choice, the more reward they received. During the scan, 4 sets of 20 trials category pair of dot patterns were used. Dataset 2 (n=136, ages 20-79) comprised resting-state scans. All data were acquired at MGH Martinos Center using identical protocols (TR=802ms, TE=20ms, flip angle=33°, voxel size=2×2×1.5mm, 85 slices). Data processing included dcm2niix conversion, fMRIPrep preprocessing, BrainDancer denoising, detrending, band-pass filtering, and confound removal using nilearn. BrainDancer denoising was done as described in (Kumar, 2021). Analysis used Seitzman's 300-ROI parcellation. We assessed task effects through single-subject modeling, voxel-wise correlations with generated task regressors, and aging effects through functional connectivity (FC) and network segregation analyses. Whole-brain FC was the mean value of the all-to-all connectivity matrix among the 300 ROIs. Network segregation was computed based on (Chan, 2014).
Results:
BrainDancer denoising reduced signal noise while preserving fMRI dynamics [Figure 1A]. In task analyses, this enhanced both correlations and anticorrelations with task regressors [Figure 1B]. In the aging dataset, BrainDancer amplified FC signals across networks [Figure 2A]. There were stronger age-related trends in FC (r=0.24, p<0.001) and network segregation (r=-0.31, p<0.0001) [Figure 2B, C, D]. There were no age-related trends in FC before BrainDancer cleaning. The trend between age and segregation was weaker (-0.18, p<0.001). These complementary findings demonstrated increased whole-brain connectivity and reduced network segregation with age, patterns that were less evident in the raw data.
Conclusions:
BrainDancer denoising effectively amplifies biological contrasts across both task-based and aging-related analyses. By reducing noise-driven false positives, the approach enhances the detection of true signal fluctuations and improves statistical power without requiring additional subjects. This method represents a significant advance in optimizing fMRI signal processing for biological relevance.
Lifespan Development:
Aging
Modeling and Analysis Methods:
Activation (eg. BOLD task-fMRI)
Exploratory Modeling and Artifact Removal 1
fMRI Connectivity and Network Modeling 2
Keywords:
Aging
Data analysis
FUNCTIONAL MRI
Machine Learning
Modeling
Statistical Methods
Other - Denoising
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Are you Internal Review Board (IRB) certified? Please note: Failure to have IRB, if applicable will lead to automatic rejection of abstract.
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Which processing packages did you use for your study?
Provide references using APA citation style.
Kumar R, Tan L, Kriegstein A, Lithen A, Polimeni JR, Mujica-Parodi LR, Strey HH. Ground-truth "resting-state" signal provides data-driven estimation and correction for scanner distortion of fMRI time-series dynamics. Neuroimage. 2021 Feb 15;227:117584. doi: 10.1016/j.neuroimage.2020.117584. Epub 2020 Dec 4. PMID: 33285328.
Chan MY, Park DC, Savalia NK, Petersen SE, Wig GS. Decreased segregation of brain systems across the healthy adult lifespan. Proc Natl Acad Sci U S A. 2014 Nov 18;111(46):E4997-5006. doi: 10.1073/pnas.1415122111. Epub 2014 Nov 3. PMID: 25368199; PMCID: PMC4246293.