A semi-supervised Teacher-Student framework for surgical tool detection and localization

TL;DR

This paper introduces a semi-supervised teacher-student framework for surgical tool detection that reduces annotation needs and improves accuracy, especially with limited labeled data, by leveraging pseudo labels and a novel loss function.

Contribution

The work presents a semi-supervised learning approach with a knowledge distillation method and a multi-class distance loss, enhancing surgical tool detection with minimal labeled data.

Findings

Achieves up to 27% improvement in mAP over fully supervised methods with 1% labeled data.

Effectively mitigates class imbalance and pseudo label bias in surgical tool detection.

Demonstrates robustness and generalization benefits of semi-supervised learning in medical imaging.

Abstract

Surgical tool detection in minimally invasive surgery is an essential part of computer-assisted interventions. Current approaches are mostly based on supervised methods which require large fully labeled data to train supervised models and suffer from pseudo label bias because of class imbalance issues. However large image datasets with bounding box annotations are often scarcely available. Semi-supervised learning (SSL) has recently emerged as a means for training large models using only a modest amount of annotated data; apart from reducing the annotation cost. SSL has also shown promise to produce models that are more robust and generalizable. Therefore, in this paper we introduce a semi-supervised learning (SSL) framework in surgical tool detection paradigm which aims to mitigate the scarcity of training data and the data imbalance through a knowledge distillation approach. In the proposed work, we train a model with labeled data which initialises the Teacher-Student joint learning, where the Student is trained on Teacher-generated pseudo labels from unlabeled data. We propose a multi-class distance with a margin based classification loss function in the region-of-interest head of the detector to effectively segregate foreground classes from background region. Our results on m2cai16-tool-locations dataset indicate the superiority of our approach on different supervised data settings (1%, 2%, 5%, 10% of annotated data) where our model achieves overall improvements of 8%, 12% and 27% in mAP (on 1% labeled data) over the state-of-the-art SSL methods and a fully supervised baseline, respectively. The code is available at https://github.com/Mansoor-at/Semi-supervised-surgical-tool-det