Co-Generation and Segmentation for Generalized Surgical Instrument Segmentation on Unlabelled Data
This work addresses the bottleneck of labeled data availability for surgical instrument segmentation in robot-assisted surgery, offering a method to enhance generalization across domains, though it is incremental as it builds on existing domain adaptation techniques.
The paper tackles the limited generalizability of deep learning-based surgical instrument segmentation methods due to reliance on labeled data by proposing a joint generation and segmentation strategy that leverages labeled data from one domain to improve segmentation on unlabeled domains, achieving consistently high mean Dice scores on both labeled and unlabeled datasets.
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