Scribble-supervised Cell Segmentation Using Multiscale Contrastive Regularization

TL;DR

This paper introduces a multiscale contrastive regularization technique to enhance scribble-supervised cell segmentation, reducing overfitting and improving accuracy to match fully supervised methods.

Contribution

The work proposes a novel multiscale contrastive regularization for scribble-supervised segmentation, addressing overfitting and bias issues inherent in small scribble annotations.

Findings

Improved segmentation performance on benchmark datasets.

Multiscale contrastive loss enhances feature separation at various scales.

Comparable results to fully supervised methods.

Abstract

Current state-of-the-art supervised deep learning-based segmentation approaches have demonstrated superior performance in medical image segmentation tasks. However, such supervised approaches require fully annotated pixel-level ground-truth labels, which are labor-intensive and time-consuming to acquire. Recently, Scribble2Label (S2L) demonstrated that using only a handful of scribbles with self-supervised learning can generate accurate segmentation results without full annotation. However, owing to the relatively small size of scribbles, the model is prone to overfit and the results may be biased to the selection of scribbles. In this work, we address this issue by employing a novel multiscale contrastive regularization term for S2L. The main idea is to extract features from intermediate layers of the neural network for contrastive loss so that structures at various scales can be effectively separated. To verify the efficacy of our method, we conducted ablation studies on well-known datasets, such as Data Science Bowl 2018 and MoNuSeg. The results show that the proposed multiscale contrastive loss is effective in improving the performance of S2L, which is comparable to that of the supervised learning segmentation method.