Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Automatic Gleason grading of H&E stained microscopic prostate images using deep convolutional neural networks

Gummeson, Anna ; Arvidsson, Ida LU ; Ohlsson, Mattias LU orcid ; Overgaard, Niels C. LU ; Krzyzanowska, Agnieszka LU ; Heyden, Anders LU orcid ; Bjartell, Anders LU and Aström, Kalle LU orcid (2017) Medical Imaging 2017: Digital Pathology 10140.
Abstract

Prostate cancer is the most diagnosed cancer in men. The diagnosis is confirmed by pathologists based on ocular inspection of prostate biopsies in order to classify them according to Gleason score. The main goal of this paper is to automate the classification using convolutional neural networks (CNNs). The introduction of CNNs has broadened the field of pattern recognition. It replaces the classical way of designing and extracting hand-made features used for classification with the substantially different strategy of letting the computer itself decide which features are of importance. For automated prostate cancer classification into the classes: Benign, Gleason grade 3, 4 and 5 we propose a CNN with small convolutional filters that has... (More)

Prostate cancer is the most diagnosed cancer in men. The diagnosis is confirmed by pathologists based on ocular inspection of prostate biopsies in order to classify them according to Gleason score. The main goal of this paper is to automate the classification using convolutional neural networks (CNNs). The introduction of CNNs has broadened the field of pattern recognition. It replaces the classical way of designing and extracting hand-made features used for classification with the substantially different strategy of letting the computer itself decide which features are of importance. For automated prostate cancer classification into the classes: Benign, Gleason grade 3, 4 and 5 we propose a CNN with small convolutional filters that has been trained from scratch using stochastic gradient descent with momentum. The input consists of microscopic images of haematoxylin and eosin stained tissue, the output is a coarse segmentation into regions of the four different classes. The dataset used consists of 213 images, each considered to be of one class only. Using four-fold cross-validation we obtained an error rate of 7.3%, which is significantly better than previous state of the art using the same dataset. Although the dataset was rather small, good results were obtained. From this we conclude that CNN is a promising method for this problem. Future work includes obtaining a larger dataset, which potentially could diminish the error margin.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; ; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
Classification, Convolutional neural networks, Deep learning, Gleason grading, Prostate cancer
host publication
Medical Imaging 2017: Digital Pathology
volume
10140
article number
101400S
publisher
SPIE
conference name
Medical Imaging 2017: Digital Pathology
conference location
Orlando, United States
conference dates
2017-02-12 - 2017-02-13
external identifiers
  • wos:000404880200026
  • scopus:85020314166
ISBN
9781510607255
DOI
10.1117/12.2253620
project
Lund University AI Research
language
English
LU publication?
yes
id
85c3a0e7-18ce-4911-951b-fd95b0f2bdea
date added to LUP
2017-08-09 12:14:47
date last changed
2024-04-14 15:12:35
@inproceedings{85c3a0e7-18ce-4911-951b-fd95b0f2bdea,
  abstract     = {{<p>Prostate cancer is the most diagnosed cancer in men. The diagnosis is confirmed by pathologists based on ocular inspection of prostate biopsies in order to classify them according to Gleason score. The main goal of this paper is to automate the classification using convolutional neural networks (CNNs). The introduction of CNNs has broadened the field of pattern recognition. It replaces the classical way of designing and extracting hand-made features used for classification with the substantially different strategy of letting the computer itself decide which features are of importance. For automated prostate cancer classification into the classes: Benign, Gleason grade 3, 4 and 5 we propose a CNN with small convolutional filters that has been trained from scratch using stochastic gradient descent with momentum. The input consists of microscopic images of haematoxylin and eosin stained tissue, the output is a coarse segmentation into regions of the four different classes. The dataset used consists of 213 images, each considered to be of one class only. Using four-fold cross-validation we obtained an error rate of 7.3%, which is significantly better than previous state of the art using the same dataset. Although the dataset was rather small, good results were obtained. From this we conclude that CNN is a promising method for this problem. Future work includes obtaining a larger dataset, which potentially could diminish the error margin.</p>}},
  author       = {{Gummeson, Anna and Arvidsson, Ida and Ohlsson, Mattias and Overgaard, Niels C. and Krzyzanowska, Agnieszka and Heyden, Anders and Bjartell, Anders and Aström, Kalle}},
  booktitle    = {{Medical Imaging 2017: Digital Pathology}},
  isbn         = {{9781510607255}},
  keywords     = {{Classification; Convolutional neural networks; Deep learning; Gleason grading; Prostate cancer}},
  language     = {{eng}},
  publisher    = {{SPIE}},
  title        = {{Automatic Gleason grading of H&E stained microscopic prostate images using deep convolutional neural networks}},
  url          = {{http://dx.doi.org/10.1117/12.2253620}},
  doi          = {{10.1117/12.2253620}},
  volume       = {{10140}},
  year         = {{2017}},
}