Separable-HoverNet and Instance-YOLO for Colon Nuclei Identification and Counting

03/01/2022
by   Chunhui Lin, et al.
0

Nuclear segmentation, classification and quantification within Haematoxylin Eosin stained histology images enables the extraction of interpretable cell-based features that can be used in downstream explainable models in computational pathology (CPath). However, automatic recognition of different nuclei is faced with a major challenge in that there are several different types of nuclei, some of them exhibiting large intraclass variability. In this work, we propose an approach that combine Separable-HoverNet and Instance-YOLOv5 to indentify colon nuclei small and unbalanced. Our approach can achieve mPQ+ 0.389 on the Segmentation and Classification-Preliminary Test Dataset and r2 0.599 on the Cellular Composition-Preliminary Test Dataset on ISBI 2022 CoNIC Challenge.

READ FULL TEXT

page 1

page 2

research
11/29/2021

CoNIC: Colon Nuclei Identification and Counting Challenge 2022

Nuclear segmentation, classification and quantification within Haematoxy...
research
03/04/2022

Cellular Segmentation and Composition in Routine Histology Images using Deep Learning

Identification and quantification of nuclei in colorectal cancer haemato...
research
02/23/2022

Nuclei panoptic segmentation and composition regression with multi-task deep neural networks

Nuclear segmentation, classification and quantification within Haematoxy...
research
02/25/2022

ciscNet – A Single-Branch Cell Instance Segmentation and Classification Network

Automated cell nucleus segmentation and classification are required to a...
research
03/11/2021

Learning with partially separable data

There are partially separable data types that make classification tasks ...
research
03/01/2022

Nuclear Segmentation and Classification Model with Imbalanced Classes for CoNiC Challenge

Nuclear segmentation and classification is an essential step for computa...
research
12/28/2022

SynCLay: Interactive Synthesis of Histology Images from Bespoke Cellular Layouts

Automated synthesis of histology images has several potential applicatio...

Please sign up or login with your details

Forgot password? Click here to reset