Related Experiment Video
Updated: Jan 3, 2026
09:55
Neuroimaging-Guided TMS–EEG for Real-Time Cortical Network Mapping
Published on: June 13, 2025
2.3K
Neural Segmentation of Seeding ROIs (sROIs) for Pre-Surgical Brain Tractography.
IEEE Transactions on Medical Imaging
|November 22, 2019
Summary
Automated segmentation of seeding regions of interest (sROIs) using multi-modal fully convolutional networks can improve neuro-surgical planning. This approach fuses anatomical T1w maps and directionally encoded color maps for efficient and accurate tractography mapping.
Area of Science:
- Neuroimaging
- Computational Neuroscience
- Medical Image Analysis
Background:
- White matter tractography is crucial for neuro-surgical planning and navigation.
- Accurate manual delineation of seeding regions of interest (sROIs) by experts is time-consuming and resource-intensive.
- There is a significant need for automated tools to streamline pre-operative neurosurgical workflows.
Purpose of the Study:
- To propose and compare multi-modal fully convolutional network architectures for automated sROI segmentation.
- To integrate anatomical information from T1w maps with directionally encoded color (DEC) maps for improved segmentation accuracy.
- To evaluate the performance of the proposed networks against state-of-the-art methods.
Main Methods:
- Development of several multi-modal fully convolutional network architectures for sROI segmentation.
- Fusion of T1w anatomical maps and DEC maps within the network architectures.
- Qualitative and quantitative validation using neuroimaging data from 75 tumor resection candidates.
- Comparison with existing state-of-the-art methods and the ISMRM 2017 traCED challenge dataset.
Main Results:
- The proposed networks demonstrated promising results in segmenting sROIs for the motor tract, arcuate fasciculus, and optic radiation.
- Favorable comparisons were achieved against state-of-the-art methods on both the tumor dataset and the ISMRM 2017 traCED challenge dataset.
- The automated segmentation approach showed potential to significantly enhance the efficiency of pre-surgical tractography mapping.
Conclusions:
- The developed multi-modal fully convolutional networks offer an effective solution for automated sROI segmentation in neurosurgery.
- This automation can significantly improve the efficiency of pre-surgical tractography mapping without compromising its quality.
- The findings suggest a potential for widespread adoption in clinical neurosurgical planning and navigation.
