Integrating Multimodal Neuroimaging for Accurate and Efficient Deep Brain Activity Simulations
Poster No:
1254
Submission Type:
Abstract Submission
Authors:
Alejandro Salinas-Medina1, Anisleidy Gonzalez-Mitjans1, Jordan DeKraker1, Paule Toussaint1, Xue Liu2, Alan Evans1
Institutions:
1McGill University, Montreal, QC, 2McGill, Montreal, QC
First Author:
Co-Author(s):
Introduction:
Simulating deep brain activity with high anatomical and functional precision requires integrating diverse multimodal neuroimaging data. Current methods often face challenges in seamlessly processing high-resolution data and maintaining topological fidelity during data scaling and interpolation. This work presents a comprehensive simulation framework designed to address these limitations and enable accurate deep brain activity simulations.
Methods:
Our framework (Figure 1) facilitates the integration of diverse modalities, including surface meshes (Figure 2B), structural connectivity data (Figure 2A), and parcellation schemes (Figure 2C), to model neural dynamics at multiple scales. Key methodological components include:
• Mapping neuroanatomical labels to surface vertices, interpolating across densities using grid-based methods (nearest-neighbor and linear interpolation) optimized for topological and categorical fidelity.
• Iterative data imputation to address missing vertex data by progressively filling gaps with mean values of neighboring vertices, preserving spatial and structural integrity.
• Rigorous validation using statistical analyses, including a chi-squared test to assess label distribution consistency between original and interpolated datasets, and normalized frequency comparisons with visual representations (bar plots) to detect discrepancies.
These methods ensure robust handling of both continuous and categorical data, enabling the creation of high-resolution, multimodal models.
• Mapping neuroanatomical labels to surface vertices, interpolating across densities using grid-based methods (nearest-neighbor and linear interpolation) optimized for topological and categorical fidelity.
• Iterative data imputation to address missing vertex data by progressively filling gaps with mean values of neighboring vertices, preserving spatial and structural integrity.
• Rigorous validation using statistical analyses, including a chi-squared test to assess label distribution consistency between original and interpolated datasets, and normalized frequency comparisons with visual representations (bar plots) to detect discrepancies.
These methods ensure robust handling of both continuous and categorical data, enabling the creation of high-resolution, multimodal models.
·Figure 1. Framework´s flow diagram
Results:
The framework demonstrated high accuracy and efficiency in integrating and processing multimodal neuroimaging data:
• Topological Accuracy: Grid-based interpolation, using the nearest-neighbor method, preserved topological fidelity, achieving a 98.7% match between interpolated and original neuroanatomical labels. This high degree of correspondence demonstrates the effectiveness of our approach in maintaining anatomical integrity during data scaling.
• Data Integrity: Chi-squared tests confirmed no significant differences (p > 0.05) in label distributions between interpolated and original datasets, further validating the accuracy of the interpolation process.
• Processing Efficiency: Compared to standard interpolation methods, our framework reduced computation time by 35% while maintaining high accuracy. This improvement in efficiency allows for faster processing of large datasets and facilitates more rapid analysis.
This combination of high accuracy and improved efficiency underscores the effectiveness of our framework for integrating and processing multimodal neuroimaging data.
• Topological Accuracy: Grid-based interpolation, using the nearest-neighbor method, preserved topological fidelity, achieving a 98.7% match between interpolated and original neuroanatomical labels. This high degree of correspondence demonstrates the effectiveness of our approach in maintaining anatomical integrity during data scaling.
• Data Integrity: Chi-squared tests confirmed no significant differences (p > 0.05) in label distributions between interpolated and original datasets, further validating the accuracy of the interpolation process.
• Processing Efficiency: Compared to standard interpolation methods, our framework reduced computation time by 35% while maintaining high accuracy. This improvement in efficiency allows for faster processing of large datasets and facilitates more rapid analysis.
This combination of high accuracy and improved efficiency underscores the effectiveness of our framework for integrating and processing multimodal neuroimaging data.
Conclusions:
Our framework provides a robust and validated approach for integrating multimodal neuroimaging data to enable high-precision deep brain activity simulations. The demonstrated consistency between original and interpolated data, validated through statistical analyses, highlights the framework's ability to create accurate high-resolution multimodal representations. Output compatibility with leading neural simulation platforms like TVB facilitates seamless integration into existing simulation workflows. While the hippocampus served as a case study, the methodology is designed to generalize across brain regions and scales, supporting diverse research in computational neuroscience.
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural) 1
Segmentation and Parcellation 2
Keywords:
Atlasing
Computational Neuroscience
Data analysis
Data Organization
Electroencephaolography (EEG)
Machine Learning
Modeling
Morphometrics
STRUCTURAL MRI
Systems
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?
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:
Provide references using APA citation style.
2. P. Sanz Leon et al. “The Virtual Brain: A simulator of primate brain network dynamics”. In:Frontiers in Neuroinformatics 7 (2013). url: https://doi.org/10.3389/fninf.2013.00010.
3. Jordan DeKraker, Roy AM Haast, et al. “Automated hippocampal unfolding for morphometry and subfield segmentation with HippUnfold”. In: eLife 11 (2022), e77945