Co-Generation and Segmentation for Generalized Surgical Instrument Segmentation on Unlabelled Data

TL;DR

This paper introduces a joint generation and segmentation approach that improves the generalization of surgical instrument segmentation models across different datasets and unlabelled data domains, addressing the limitations of existing deep learning methods.

Contribution

The paper proposes a novel joint generation and segmentation strategy that leverages labelled data from one domain to improve segmentation in unlabelled domains, enhancing generalization capabilities.

Findings

High mean Dice scores on both labelled and unlabelled datasets

Outperforms state-of-the-art methods in generalization

Effective on real robot-assisted surgery videos

Abstract

Surgical instrument segmentation for robot-assisted surgery is needed for accurate instrument tracking and augmented reality overlays. Therefore, the topic has been the subject of a number of recent papers in the CAI community. Deep learning-based methods have shown state-of-the-art performance for surgical instrument segmentation, but their results depend on labelled data. However, labelled surgical data is of limited availability and is a bottleneck in surgical translation of these methods. In this paper, we demonstrate the limited generalizability of these methods on different datasets, including human robot-assisted surgeries. We then propose a novel joint generation and segmentation strategy to learn a segmentation model with better generalization capability to domains that have no labelled data. The method leverages the availability of labelled data in a different domain. The generator does the domain translation from the labelled domain to the unlabelled domain and simultaneously, the segmentation model learns using the generated data while regularizing the generative model. We compared our method with state-of-the-art methods and showed its generalizability on publicly available datasets and on our own recorded video frames from robot-assisted prostatectomies. Our method shows consistently high mean Dice scores on both labelled and unlabelled domains when data is available only for one of the domains. *M. Kalia and T. Aleef contributed equally to the manuscript