Simultaneous Semantic and Instance Segmentation for Colon Nuclei Identification and Counting

TL;DR

This paper introduces a combined semantic and instance segmentation framework for automated colon nuclei identification and counting, improving accuracy in histology image analysis for pathology applications.

Contribution

The paper presents a novel ensemble approach combining HoverNet and Cascade Mask-RCNN with custom NMS for improved nuclei segmentation and classification.

Findings

Outperforms baseline methods significantly

Effective ensemble of semantic and instance segmentation models

Accurate nuclei counting and classification in histology images

Abstract

We address the problem of automated nuclear segmentation, classification, and quantification from Haematoxylin and Eosin stained histology images, which is of great relevance for several downstream computational pathology applications. In this work, we present a solution framed as a simultaneous semantic and instance segmentation framework. Our solution is part of the Colon Nuclei Identification and Counting (CoNIC) Challenge. We first train a semantic and instance segmentation model separately. Our framework uses as backbone HoverNet and Cascade Mask-RCNN models. We then ensemble the results with a custom Non-Maximum Suppression embedding (NMS). In our framework, the semantic model computes a class prediction for the cells whilst the instance model provides a refined segmentation. We demonstrate, through our experimental results, that our model outperforms the provided baselines by a large margin.