Affine-Consistent Transformer for Multi-Class Cell Nuclei Detection

TL;DR

This paper introduces an Affine-Consistent Transformer that directly predicts cell nuclei locations in histopathology images, utilizing a collaborative global-local training approach with an adaptive affine transformation module to improve detection accuracy.

Contribution

The paper presents a novel transformer-based framework with an adaptive affine transformation module for improved multi-class cell nuclei detection in histopathology images.

Findings

Outperforms state-of-the-art methods on multiple benchmarks.

Effectively handles diverse cell morphologies and distributions.

Utilizes a collaborative training scheme with global and local networks.

Abstract

Multi-class cell nuclei detection is a fundamental prerequisite in the diagnosis of histopathology. It is critical to efficiently locate and identify cells with diverse morphology and distributions in digital pathological images. Most existing methods take complex intermediate representations as learning targets and rely on inflexible post-refinements while paying less attention to various cell density and fields of view. In this paper, we propose a novel Affine-Consistent Transformer (AC-Former), which directly yields a sequence of nucleus positions and is trained collaboratively through two sub-networks, a global and a local network. The local branch learns to infer distorted input images of smaller scales while the global network outputs the large-scale predictions as extra supervision signals. We further introduce an Adaptive Affine Transformer (AAT) module, which can automatically learn the key spatial transformations to warp original images for local network training. The AAT module works by learning to capture the transformed image regions that are more valuable for training the model. Experimental results demonstrate that the proposed method significantly outperforms existing state-of-the-art algorithms on various benchmarks.