First Investigation of Deep Learning for Intraoperative Gauze Segmentation in Minimally Invasive Abdominal Surgery

2026-07-31

AI Summary

This study addresses the critical challenge of intraoperative gauze retention, which can lead to severe complications, yet remains difficult to detect accurately in real surgical settings due to scarce annotated data and interference from blood. For the first time, this work systematically investigates semantic segmentation of three types of gauze in real robot-assisted laparoscopic surgery videos, employing CNN, Transformer, and hybrid architectures. To mitigate the scarcity of high-quality annotations, the authors innovatively introduce automatically generated suboptimal labels via tracking-based pseudo-labeling. Experimental results demonstrate that models trained on real surgical data effectively handle blood occlusion, and the proposed automatic labeling strategy significantly enhances segmentation performance, offering a robust foundation for foreign object detection in robot-assisted surgery.

Abstract

Surgical gauze is an essential part of surgical procedures, primarily used for controlling bleeding and absorbing bodily fluids. The post-surgical retention of gauze can lead to serious complications and necessitate additional surgery for its removal. Despite the clinical significance, research on gauze segmentation using real-world surgical data remains underexplored, owing in part to the scarcity of annotated datasets. In this work, we investigate the use of deep learning methods for gauze segmentation in robot-assisted minimally invasive abdominal surgeries, utilizing an in-house surgical dataset prepared at a university hospital. The training data reflects realistic surgical settings and captures extensive diversity in spatial, morphological, and visual attributes across three different gauze categories. We evaluate several widely used segmentation architectures, including CNN-based, transformer-based, and hybrid architectures, to establish a proof-of-concept for gauze segmentation in a realistic clinical setting. In addition, we investigate the influence of sub-optimally annotated, auto-tracked segmentation masks as a strategy to address data scarcity and improve performance. Our results demonstrate the efficacy of real-world training data in countering the main challenge reported by prior works, the trade-off between blood presence and gauze detection. The incorporation of auto-tracked annotations yields performance enhancements, particularly in generic surgical scenarios. The integration of effective segmentation approaches can benefit robot-guided surgical procedures and various downstream applications by providing precise delineation of foreign objects, thereby enhancing patient safety and surgical outcomes.

Problem

• gauze segmentation
• minimally invasive surgery
• surgical safety
• data scarcity
• foreign object retention

Innovation

• gauze segmentation
• deep learning
• minimally invasive surgery
• auto-tracked annotations
• surgical safety