Data-Efficient Surgical Phase Segmentation in Small-Incision Cataract Surgery: A Controlled Study of Vision Foundation Models

TL;DR

This study demonstrates that modern vision foundation models significantly enhance data-efficient surgical phase segmentation in small-incision cataract surgery, especially in low-label settings, by leveraging transferability and lightweight adaptation.

Contribution

It provides a controlled comparison showing foundation models outperform traditional encoders in surgical video segmentation and offers practical insights for low-label medical video applications.

Findings

DINOv3 ViT-7B achieves 83.4% accuracy in segmentation.

Foundation models improve performance over supervised encoders.

Lightweight adaptation benefits transfer learning in surgical videos.

Abstract

Surgical phase segmentation is central to computer-assisted surgery, yet robust models remain difficult to develop when labeled surgical videos are scarce. We study data-efficient phase segmentation for manual small-incision cataract surgery (SICS) through a controlled comparison of visual representations. To isolate representation quality, we pair each visual encoder with the same temporal model (MS-TCN++) under identical training and evaluation settings on SICS-155 (19 phases). We compare supervised encoders (ResNet-50, I3D) against large self-supervised foundation models (DINOv3, V-JEPA2), and use a cached-feature pipeline that decouples expensive visual encoding from lightweight temporal learning. Foundation-model features improve segmentation performance in this setup, with DINOv3 ViT-7B achieving the best overall results (83.4% accuracy, 87.0 edit score). We further examine cataract-domain transfer using unlabeled videos and lightweight adaptation, and analyze when it helps or hurts. Overall, the study indicates strong transferability of modern vision foundation models to surgical workflow understanding and provides practical guidance for low-label medical video settings. The project website is available at: https://sl2005.github.io/DataEfficient-sics-phase-seg/