Automatic segmentation of colorectal liver metastases for ultrasound-based navigated resection

2025-11-07

AI Summary

Intraoperative ultrasound (iUS) for colorectal liver metastases (CRLM) suffers from low segmentation accuracy, high noise levels, and strong operator dependence. Method: We propose a registration-free, near-real-time automatic segmentation method based on an optimized 3D U-Net architecture derived from nnU-Net, specifically designed to process tracked 3D iUS sequences directly; the model is integrated into the 3D Slicer platform for intraoperative navigation. Contribution/Results: Prospective intraoperative evaluation yielded a median Dice coefficient of 0.74 and inference speed ~4× faster than semi-automatic methods, significantly reducing manual intervention and operator dependency. Our key innovation lies in the first integration of region cropping with dynamic temporal modeling of iUS sequences, enabling robust, low-latency, clinically deployable CRLM segmentation—thereby providing reliable technical support for achieving negative surgical margins.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Large Multimodal Models (LMMs)
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Introduction: Accurate intraoperative delineation of colorectal liver metastases (CRLM) is crucial for achieving negative resection margins but remains challenging using intraoperative ultrasound (iUS) due to low contrast, noise, and operator dependency. Automated segmentation could enhance precision and efficiency in ultrasound-based navigation workflows. Methods: Eighty-five tracked 3D iUS volumes from 85 CRLM patients were used to train and evaluate a 3D U-Net implemented via the nnU-Net framework. Two variants were compared: one trained on full iUS volumes and another on cropped regions around tumors. Segmentation accuracy was assessed using Dice Similarity Coefficient (DSC), Hausdorff Distance (HDist.), and Relative Volume Difference (RVD) on retrospective and prospective datasets. The workflow was integrated into 3D Slicer for real-time intraoperative use. Results: The cropped-volume model significantly outperformed the full-volume model across all metrics (AUC-ROC = 0.898 vs 0.718). It achieved median DSC = 0.74, recall = 0.79, and HDist. = 17.1 mm comparable to semi-automatic segmentation but with ~4x faster execution (~ 1 min). Prospective intraoperative testing confirmed robust and consistent performance, with clinically acceptable accuracy for real-time surgical guidance. Conclusion: Automatic 3D segmentation of CRLM in iUS using a cropped 3D U-Net provides reliable, near real-time results with minimal operator input. The method enables efficient, registration-free ultrasound-based navigation for hepatic surgery, approaching expert-level accuracy while substantially reducing manual workload and procedure time.

Problem

• Automated segmentation of colorectal liver metastases in ultrasound
• Enhancing precision in intraoperative ultrasound navigation workflows
• Reducing operator dependency and manual workload in hepatic surgery

Innovation

• Cropped 3D U-Net for liver metastasis segmentation
• nnU-Net framework enabling automatic ultrasound analysis
• Real-time integration in 3D Slicer surgical navigation