Colon Nuclei Instance Segmentation using a Probabilistic Two-Stage Detector

TL;DR

This paper introduces SegCenterNet2, an enhanced deep learning model for nuclei instance segmentation in histopathology images, demonstrating superior performance over Mask R-CNN in the CoNIC challenge.

Contribution

The paper presents a novel adaptation of CenterNet2 for simultaneous object detection and instance segmentation in medical images.

Findings

SegCenterNet2 outperforms Mask R-CNN in CoNIC challenge metrics.

The model effectively segments nuclei in histopathology images.

Deep learning can assist pathologists in cancer diagnosis.

Abstract

Cancer is one of the leading causes of death in the developed world. Cancer diagnosis is performed through the microscopic analysis of a sample of suspicious tissue. This process is time consuming and error prone, but Deep Learning models could be helpful for pathologists during cancer diagnosis. We propose to change the CenterNet2 object detection model to also perform instance segmentation, which we call SegCenterNet2. We train SegCenterNet2 in the CoNIC challenge dataset and show that it performs better than Mask R-CNN in the competition metrics.