Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

UMambaAdj: Advancing GTV Segmentation for Head and Neck Cancer in MRI-Guided RT with UMamba and nnU-Net ResEnc Planner

Jintao Ren, Kim Hochreuter, Jesper Folsted Kallehauge, Stine Sofia Korreman

TL;DR

This approach demonstrates potential for more precise tumor delineation in MRI-guided adaptive radiotherapy, ultimately improving treatment outcomes for HNC patients.

Abstract

Magnetic Resonance Imaging (MRI) plays a crucial role in MRI-guided adaptive radiotherapy for head and neck cancer (HNC) due to its superior soft-tissue contrast. However, accurately segmenting the gross tumor volume (GTV), which includes both the primary tumor (GTVp) and lymph nodes (GTVn), remains challenging. Recently, two deep learning segmentation innovations have shown great promise: UMamba, which effectively captures long-range dependencies, and the nnU-Net Residual Encoder (ResEnc), which enhances feature extraction through multistage residual blocks. In this study, we integrate these strengths into a novel approach, termed 'UMambaAdj'. Our proposed method was evaluated on the HNTS-MRG 2024 challenge test set using pre-RT T2-weighted MRI images, achieving an aggregated Dice Similarity Coefficient (DSCagg) of 0.751 for GTVp and 0.842 for GTVn, with a mean DSCagg of 0.796. This approach demonstrates potential for more precise tumor delineation in MRI-guided adaptive radiotherapy, ultimately improving treatment outcomes for HNC patients. Team: DCPT-Stine's group.

UMambaAdj: Advancing GTV Segmentation for Head and Neck Cancer in MRI-Guided RT with UMamba and nnU-Net ResEnc Planner

TL;DR

This approach demonstrates potential for more precise tumor delineation in MRI-guided adaptive radiotherapy, ultimately improving treatment outcomes for HNC patients.

Abstract

Magnetic Resonance Imaging (MRI) plays a crucial role in MRI-guided adaptive radiotherapy for head and neck cancer (HNC) due to its superior soft-tissue contrast. However, accurately segmenting the gross tumor volume (GTV), which includes both the primary tumor (GTVp) and lymph nodes (GTVn), remains challenging. Recently, two deep learning segmentation innovations have shown great promise: UMamba, which effectively captures long-range dependencies, and the nnU-Net Residual Encoder (ResEnc), which enhances feature extraction through multistage residual blocks. In this study, we integrate these strengths into a novel approach, termed 'UMambaAdj'. Our proposed method was evaluated on the HNTS-MRG 2024 challenge test set using pre-RT T2-weighted MRI images, achieving an aggregated Dice Similarity Coefficient (DSCagg) of 0.751 for GTVp and 0.842 for GTVn, with a mean DSCagg of 0.796. This approach demonstrates potential for more precise tumor delineation in MRI-guided adaptive radiotherapy, ultimately improving treatment outcomes for HNC patients. Team: DCPT-Stine's group.

Paper Structure

This paper contains 14 sections, 2 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Material and Methods
  3. Data
  4. Network Architecture
  5. Mamba layer
  6. Training Parameters
  7. Evaluation
  8. System environment
  9. Results
  10. Cross-Validation performance
  11. Comparision of nnU-Net, ResEnc, UmambaEnc and UmambaAdj
  12. Final test score
  13. Discussion
  14. Conclusion

Figures (2)

  • Figure 1: (a) Overview of the proposed UMamba adjustment (UMambaAdj) network architecture. (b) Details of the Mamba layer.
  • Figure 2: Two patients (a and b) were selected for illustration. For each patient, the first row (left to right) displays the original T2-weighted MRI image, the ground truth overlaid on the image, and the segmentation results from all compared methods overlaid on the image. The second row shows the 3D renderings of the delineations and segmentations. Red represents the GTVp, and Green represents the GTVn. Segmentation metrics, including DSC and HD, are shown on the rendering subfigures. A yellow arrow in patient (b) indicates a nearly invisible small false-positive GTVp segmentation predicted by UMambaEnc.