Cross-Dataset Adaptation for Instrument Classification in Cataract Surgery Videos

TL;DR

This paper introduces the Barlow Adaptor, an unsupervised domain adaptation method with a novel loss function, BFAL, that improves instrument classification across different cataract surgery datasets without requiring labeled data from target domains.

Contribution

The paper proposes a new end-to-end UDA method and a novel loss function, BFAL, to address domain shift in cataract surgery videos, outperforming existing methods.

Findings

Outperforms state-of-the-art UDA methods by 6% on cataract datasets

BFAL reduces feature redundancy and improves cross-dataset generalization

Effective for instrument classification without labeled target data

Abstract

Surgical tool presence detection is an important part of the intra-operative and post-operative analysis of a surgery. State-of-the-art models, which perform this task well on a particular dataset, however, perform poorly when tested on another dataset. This occurs due to a significant domain shift between the datasets resulting from the use of different tools, sensors, data resolution etc. In this paper, we highlight this domain shift in the commonly performed cataract surgery and propose a novel end-to-end Unsupervised Domain Adaptation (UDA) method called the Barlow Adaptor that addresses the problem of distribution shift without requiring any labels from another domain. In addition, we introduce a novel loss called the Barlow Feature Alignment Loss (BFAL) which aligns features across different domains while reducing redundancy and the need for higher batch sizes, thus improving cross-dataset performance. The use of BFAL is a novel approach to address the challenge of domain shift in cataract surgery data. Extensive experiments are conducted on two cataract surgery datasets and it is shown that the proposed method outperforms the state-of-the-art UDA methods by 6%. The code can be found at https://github.com/JayParanjape/Barlow-Adaptor