Whole slide image classification via iterative patch labelling

Chaoyi Zhang; Yang Song; Donghao Zhang; Sidong Liu; Mei Chen; Weidong Cai

doi:10.1109/ICIP.2018.8451551

Whole slide image classification via iterative patch labelling

Chaoyi Zhang, Yang Song, Donghao Zhang, Sidong Liu, Mei Chen, Weidong Cai

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

9 Citations (Scopus)

Abstract

Brain tumor can be a fatal disease in the world. With the aim of improving survival rates, many computerized algorithms have been proposed to assist the pathologists to make a diagnosis' using Whole Slide Pathology Images (WSI). Most methods focus on performing patch-level classification and aggregating the patch-level results to obtain the image classification. Since not all patches carry diagnostic information, it is thus important for our algorithm to recognize discriminative and non-discriminative patches. In this study, we propose an iterative patch labelling algorithm based on the Convolutional Neural Network (CNN), with a well-designed thresholding scheme, a training policy and a novel discriminative model architecture, to distinguish patches and use the discriminative ones to achieve WSI -classification. Our method is evaluated on the MICCAI 2015 Challenge Dataset, and shows a large improvement over the baseline approaches.

Original language	English
Title of host publication	Proceedings of 2018 IEEE International Conference on Image Processing
Editors	Christophoros Nikou, Kostas Plataniotis
Place of Publication	USA
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	1408-1412
Number of pages	5
Edition	1st
ISBN (Electronic)	9781479970612
DOIs	https://doi.org/10.1109/ICIP.2018.8451551
Publication status	Published - 2018
Externally published	Yes
Event	IEEE International Conference on Image Processing 2018 - Athens, Greece Duration: 7 Oct 2018 → 10 Oct 2018 Conference number: 25th https://2018.ieeeicip.org/ https://ieeexplore.ieee.org/xpl/conhome/8436606/proceeding (Proceedings)

Publication series

Name	Proceedings - International Conference on Image Processing, ICIP
ISSN (Print)	1522-4880

Conference

Conference	IEEE International Conference on Image Processing 2018
Abbreviated title	ICIP 2018
Country/Territory	Greece
City	Athens
Period	7/10/18 → 10/10/18
Internet address	https://2018.ieeeicip.org/ https://ieeexplore.ieee.org/xpl/conhome/8436606/proceeding (Proceedings)

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

Brain cancer
Classification
Discriminative patches
Iterative patch labelling
WSI

Access to Document

10.1109/ICIP.2018.8451551

Cite this

Zhang, C., Song, Y., Zhang, D., Liu, S., Chen, M., & Cai, W. (2018). Whole slide image classification via iterative patch labelling. In C. Nikou, & K. Plataniotis (Eds.), Proceedings of 2018 IEEE International Conference on Image Processing (1st ed., pp. 1408-1412). (Proceedings - International Conference on Image Processing, ICIP). IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICIP.2018.8451551

Zhang, Chaoyi ; Song, Yang ; Zhang, Donghao et al. / Whole slide image classification via iterative patch labelling. Proceedings of 2018 IEEE International Conference on Image Processing. editor / Christophoros Nikou ; Kostas Plataniotis. 1st. ed. USA : IEEE, Institute of Electrical and Electronics Engineers, 2018. pp. 1408-1412 (Proceedings - International Conference on Image Processing, ICIP).

@inproceedings{92b76b1007b948ae8f2b5d41a2af6f26,

title = "Whole slide image classification via iterative patch labelling",

abstract = "Brain tumor can be a fatal disease in the world. With the aim of improving survival rates, many computerized algorithms have been proposed to assist the pathologists to make a diagnosis' using Whole Slide Pathology Images (WSI). Most methods focus on performing patch-level classification and aggregating the patch-level results to obtain the image classification. Since not all patches carry diagnostic information, it is thus important for our algorithm to recognize discriminative and non-discriminative patches. In this study, we propose an iterative patch labelling algorithm based on the Convolutional Neural Network (CNN), with a well-designed thresholding scheme, a training policy and a novel discriminative model architecture, to distinguish patches and use the discriminative ones to achieve WSI -classification. Our method is evaluated on the MICCAI 2015 Challenge Dataset, and shows a large improvement over the baseline approaches.",

keywords = "Brain cancer, Classification, Discriminative patches, Iterative patch labelling, WSI",

author = "Chaoyi Zhang and Yang Song and Donghao Zhang and Sidong Liu and Mei Chen and Weidong Cai",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; IEEE International Conference on Image Processing 2018, ICIP 2018 ; Conference date: 07-10-2018 Through 10-10-2018",

year = "2018",

doi = "10.1109/ICIP.2018.8451551",

language = "English",

series = "Proceedings - International Conference on Image Processing, ICIP",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

pages = "1408--1412",

editor = "Christophoros Nikou and Kostas Plataniotis",

booktitle = "Proceedings of 2018 IEEE International Conference on Image Processing",

address = "United States of America",

edition = "1st",

url = "https://2018.ieeeicip.org/, https://ieeexplore.ieee.org/xpl/conhome/8436606/proceeding",

}

Zhang, C, Song, Y, Zhang, D, Liu, S, Chen, M & Cai, W 2018, Whole slide image classification via iterative patch labelling. in C Nikou & K Plataniotis (eds), Proceedings of 2018 IEEE International Conference on Image Processing. 1st edn, Proceedings - International Conference on Image Processing, ICIP, IEEE, Institute of Electrical and Electronics Engineers, USA, pp. 1408-1412, IEEE International Conference on Image Processing 2018, Athens, Greece, 7/10/18. https://doi.org/10.1109/ICIP.2018.8451551

Whole slide image classification via iterative patch labelling. / Zhang, Chaoyi; Song, Yang; Zhang, Donghao et al.
Proceedings of 2018 IEEE International Conference on Image Processing. ed. / Christophoros Nikou; Kostas Plataniotis. 1st. ed. USA: IEEE, Institute of Electrical and Electronics Engineers, 2018. p. 1408-1412 (Proceedings - International Conference on Image Processing, ICIP).

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - Whole slide image classification via iterative patch labelling

AU - Zhang, Chaoyi

AU - Song, Yang

AU - Zhang, Donghao

AU - Liu, Sidong

AU - Chen, Mei

AU - Cai, Weidong

N1 - Conference code: 25th

PY - 2018

Y1 - 2018

N2 - Brain tumor can be a fatal disease in the world. With the aim of improving survival rates, many computerized algorithms have been proposed to assist the pathologists to make a diagnosis' using Whole Slide Pathology Images (WSI). Most methods focus on performing patch-level classification and aggregating the patch-level results to obtain the image classification. Since not all patches carry diagnostic information, it is thus important for our algorithm to recognize discriminative and non-discriminative patches. In this study, we propose an iterative patch labelling algorithm based on the Convolutional Neural Network (CNN), with a well-designed thresholding scheme, a training policy and a novel discriminative model architecture, to distinguish patches and use the discriminative ones to achieve WSI -classification. Our method is evaluated on the MICCAI 2015 Challenge Dataset, and shows a large improvement over the baseline approaches.

AB - Brain tumor can be a fatal disease in the world. With the aim of improving survival rates, many computerized algorithms have been proposed to assist the pathologists to make a diagnosis' using Whole Slide Pathology Images (WSI). Most methods focus on performing patch-level classification and aggregating the patch-level results to obtain the image classification. Since not all patches carry diagnostic information, it is thus important for our algorithm to recognize discriminative and non-discriminative patches. In this study, we propose an iterative patch labelling algorithm based on the Convolutional Neural Network (CNN), with a well-designed thresholding scheme, a training policy and a novel discriminative model architecture, to distinguish patches and use the discriminative ones to achieve WSI -classification. Our method is evaluated on the MICCAI 2015 Challenge Dataset, and shows a large improvement over the baseline approaches.

KW - Brain cancer

KW - Classification

KW - Discriminative patches

KW - Iterative patch labelling

KW - WSI

UR - https://www.scopus.com/pages/publications/85062918312

U2 - 10.1109/ICIP.2018.8451551

DO - 10.1109/ICIP.2018.8451551

M3 - Conference Paper

AN - SCOPUS:85062918312

T3 - Proceedings - International Conference on Image Processing, ICIP

SP - 1408

EP - 1412

BT - Proceedings of 2018 IEEE International Conference on Image Processing

A2 - Nikou, Christophoros

A2 - Plataniotis, Kostas

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - USA

T2 - IEEE International Conference on Image Processing 2018

Y2 - 7 October 2018 through 10 October 2018

ER -

Zhang C, Song Y, Zhang D, Liu S, Chen M, Cai W. Whole slide image classification via iterative patch labelling. In Nikou C, Plataniotis K, editors, Proceedings of 2018 IEEE International Conference on Image Processing. 1st ed. USA: IEEE, Institute of Electrical and Electronics Engineers. 2018. p. 1408-1412. (Proceedings - International Conference on Image Processing, ICIP). doi: 10.1109/ICIP.2018.8451551

Whole slide image classification via iterative patch labelling

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Cite this