CoRe-DA: Contrastive Regression for Unsupervised Domain Adaptation in Surgical Skill Assessment

TL;DR

This paper introduces CoRe-DA, a contrastive regression framework for unsupervised domain adaptation in surgical skill assessment, improving cross-domain generalization without labeled target data.

Contribution

It presents the first benchmark for UDA in SSA regression and proposes a novel contrastive regression method that learns domain-invariant representations.

Findings

CoRe-DA outperforms state-of-the-art methods in cross-domain SSA tasks.

Achieves Spearman Correlation of 0.46 and 0.41 on different target datasets.

Enables scalable and reliable surgical skill assessment across diverse settings.

Abstract

Vision-based surgical skill assessment (SSA) enables objective and scalable evaluation of operative performance. Progress in this field is constrained by the high cost and time demands for manual annotation of quantitative skill scores, as well as the poor generalization of existing regression models to new surgical tasks and environments. Meanwhile, appreciable volumes of unlabeled video data are now available, motivating the development of unsupervised domain adaptation (UDA) methods for SSA. We introduce the first benchmark for UDA in SSA regression, spanning four datasets across dry-lab and clinical settings as well as open and robotic surgery. We evaluate eight representative models under challenging domain shifts and propose CoRe-DA, a novel contrastive regression-based adaptation framework. Our method learns domain-invariant representations through relative-score supervision and target-domain self-training. Comprehensive experiments across two UDA settings show that CoRe-DA is superior to state-of-the-art methods, achieving Spearman Correlation Coefficients of 0.46 and 0.41 on dry-lab and clinical target datasets, respectively, without using any labeled target data for training. Overall, CoRe-DA enables scalable SSA with reliable cross-domain generalization, where existing methods underperform. Our code and datasets will be released at https://github.com/anastadimi/CoRe-DA.