Automated Pediatric Brain Tumor Imaging Assessment Tool from CBTN: Enhancing Suprasellar Region Inclusion and Managing Limited Data with Deep Learning
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Fully-automatic skull-stripping and tumor segmentation are crucial for monitoring pediatric brain tumors (PBT). Current methods, however, often lack generalizability, particularly for rare tumors in the sellar/suprasellar regions and when applied to real-world clinical data in limited data scenarios. To address these challenges, we propose AI-driven techniques for skull-stripping and tumor segmentation.
Methods
Multi-institutional, multi-parametric MRI scans from 527 pediatric patients (n=336 for skull-stripping, n=489 for tumor segmentation) with various PBT histologies were processed to train separate nnU-Net-based deep learning models for skull-stripping, whole tumor (WT), and enhancing tumor (ET) segmentation. These models utilized single (T2/FLAIR) or multiple (T1-Gd and T2/FLAIR) input imaging sequences. Performance was evaluated using Dice scores, sensitivity, and 95% Hausdorff distances. Statistical comparisons included paired or unpaired two-sample t-tests and Pearson’s correlation coefficient based on Dice scores from different models and PBT histologies.
Results
Dice scores for the skull-stripping models for whole brain and sellar/suprasellar region segmentation were 0.98±0.01 (median 0.98) for both multi- and single-parametric models, with significant Pearson’s correlation coefficient between single- and multi-parametric Dice scores (r > 0.80; p<0.05 for all). WT Dice scores for single-input tumor segmentation models were 0.84±0.17 (median=0.90) for T2 and 0.82±0.19 (median=0.89) for FLAIR inputs. ET Dice scores were 0.65±0.35 (median=0.79) for T1-Gd+FLAIR and 0.64±0.36 (median=0.79) for T1-Gd+T2 inputs.
Conclusion
Our skull-stripping models demonstrate excellent performance and include sellar/suprasellar regions, using single- or multi-parametric inputs. Additionally, our automated tumor segmentation models can reliably delineate whole lesions and enhancing tumor regions, adapting to MRI sessions with missing sequences in limited data context.
Brief key points:
Deep learning models for skull-stripping, including the sellar/suprasellar regions, demonstrate robustness across various pediatric brain tumor histologies.
The automated brain tumor segmentation models perform reliably even in limited data scenarios.
Importance of the Study
We present robust skull-stripping models that work with single- and multi-parametric MR images and include the sellar-suprasellar regions in the extracted brain tissue. Since ∼10% of the pediatric brain tumors originate in the sellar/suprasellar region, including the deep-seated regions within the extracted brain tissue makes these models generalizable for a wider range of tumor histologies. We also present two tumor segmentation models, one for segmenting whole tumor using T2/FLAIR images, and another for segmenting enhancing tumor region using T1-Gd and T2/FLAIR images. These models demonstrate excellent performance with limited input. Both the skull-stripping and tumor segmentation models work with one- or two-input MRI sequences, making them useful in cases where multi-parametric images are not available – especially in real-world clinical scenarios. These models help to address the issue of missing data, making it possible to include subjects for longitudinal assessment and monitoring treatment response, which would have otherwise been excluded.
