Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Weakly Supervised Spatial Implicit Neural Representation Learning for 3D MRI-Ultrasound Deformable Image Registration in HDR Prostate Brachytherapy

Jing Wang, Ruirui Liu, Yu Lei, Michael J. Baine, Tian Liu, Yang Lei

TL;DR

The paper tackles the challenge of registering pre-treatment MRI to intraoperative ultrasound for HDR prostate brachytherapy. It introduces a weakly supervised spatial implicit neural representation (SINR) that models deformations as continuous spatial functions using an MLP and a stationary velocity field, guided by sparse surface priors from MRI and US contours. On public and institutional datasets, the method achieves high prostate registration accuracy (e.g., $DSC$ up to $0.93$, $MSD$ down to $0.87$ mm) and operates in near real-time (≈$3$ s per case on a RTX A6000), while facing reduced performance for bladder/rectum due to limited US field of view. This approach reduces reliance on dense voxel-wise matching, offers biologically plausible deformations, and has the potential to enhance real-time guidance and dose delivery in HDR prostate brachytherapy.

Abstract

Purpose: Accurate 3D MRI-ultrasound (US) deformable registration is critical for real-time guidance in high-dose-rate (HDR) prostate brachytherapy. We present a weakly supervised spatial implicit neural representation (SINR) method to address modality differences and pelvic anatomy challenges. Methods: The framework uses sparse surface supervision from MRI/US segmentations instead of dense intensity matching. SINR models deformations as continuous spatial functions, with patient-specific surface priors guiding a stationary velocity field for biologically plausible deformations. Validation included 20 public Prostate-MRI-US-Biopsy cases and 10 institutional HDR cases, evaluated via Dice similarity coefficient (DSC), mean surface distance (MSD), and 95% Hausdorff distance (HD95). Results: The proposed method achieved robust registration. For the public dataset, prostate DSC was $0.93 \pm 0.05$, MSD $0.87 \pm 0.10$ mm, and HD95 $1.58 \pm 0.37$ mm. For the institutional dataset, prostate CTV achieved DSC $0.88 \pm 0.09$, MSD $1.21 \pm 0.38$ mm, and HD95 $2.09 \pm 1.48$ mm. Bladder and rectum performance was lower due to ultrasound's limited field of view. Visual assessments confirmed accurate alignment with minimal discrepancies. Conclusion: This study introduces a novel weakly supervised SINR-based approach for 3D MRI-US deformable registration. By leveraging sparse surface supervision and spatial priors, it achieves accurate, robust, and computationally efficient registration, enhancing real-time image guidance in HDR prostate brachytherapy and improving treatment precision.

Weakly Supervised Spatial Implicit Neural Representation Learning for 3D MRI-Ultrasound Deformable Image Registration in HDR Prostate Brachytherapy

TL;DR

The paper tackles the challenge of registering pre-treatment MRI to intraoperative ultrasound for HDR prostate brachytherapy. It introduces a weakly supervised spatial implicit neural representation (SINR) that models deformations as continuous spatial functions using an MLP and a stationary velocity field, guided by sparse surface priors from MRI and US contours. On public and institutional datasets, the method achieves high prostate registration accuracy (e.g., up to , down to mm) and operates in near real-time (≈ s per case on a RTX A6000), while facing reduced performance for bladder/rectum due to limited US field of view. This approach reduces reliance on dense voxel-wise matching, offers biologically plausible deformations, and has the potential to enhance real-time guidance and dose delivery in HDR prostate brachytherapy.

Abstract

Purpose: Accurate 3D MRI-ultrasound (US) deformable registration is critical for real-time guidance in high-dose-rate (HDR) prostate brachytherapy. We present a weakly supervised spatial implicit neural representation (SINR) method to address modality differences and pelvic anatomy challenges. Methods: The framework uses sparse surface supervision from MRI/US segmentations instead of dense intensity matching. SINR models deformations as continuous spatial functions, with patient-specific surface priors guiding a stationary velocity field for biologically plausible deformations. Validation included 20 public Prostate-MRI-US-Biopsy cases and 10 institutional HDR cases, evaluated via Dice similarity coefficient (DSC), mean surface distance (MSD), and 95% Hausdorff distance (HD95). Results: The proposed method achieved robust registration. For the public dataset, prostate DSC was , MSD mm, and HD95 mm. For the institutional dataset, prostate CTV achieved DSC , MSD mm, and HD95 mm. Bladder and rectum performance was lower due to ultrasound's limited field of view. Visual assessments confirmed accurate alignment with minimal discrepancies. Conclusion: This study introduces a novel weakly supervised SINR-based approach for 3D MRI-US deformable registration. By leveraging sparse surface supervision and spatial priors, it achieves accurate, robust, and computationally efficient registration, enhancing real-time image guidance in HDR prostate brachytherapy and improving treatment precision.

Paper Structure

This paper contains 9 sections, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Materials and Methods
  3. Workflow Overview
  4. Dataset and Imaging Protocols
  5. Weakly Supervised SINR for Deformation Registration
  6. Evaluation Metrics
  7. Results
  8. Discussion
  9. Conclusion

Figures (3)

  • Figure 1: Flowchart of the proposed weakly supervised spatial implicit neural representation learning-based MRI-US deformable image registration method.
  • Figure 2: Visual results from the public database: (a) Axial MRI with prostate contour overlay; (b) TRUS with prostate contour; (c) Deformed MRI aligned with TRUS, with propagated prostate contour; (d) Fused discrepancy maps (yellow/blue) between deformed MRI and TRUS.
  • Figure 3: Visual results from the institutional database: (a) Axial MRI with prostate contour overlay; (b) Transrectal US with prostate contour; (c) Deformed MRI aligned with US, with propagated prostate contour; (d) Fused discrepancy maps (yellow/blue) between deformed MRI and transrectal US.