DeepFocus: Detection of out-of-focus regions in whole slide digital images using deep learning.

The development of whole slide scanners has revolutionized the field of digital pathology. Unfortunately, whole slide scanners often produce images with out-of-focus/blurry areas that limit the amount of tissue available for a pathologist to make accurate diagnosis/prognosis. Moreover, these artifac...

Full description

Bibliographic Details
Main Authors: Caglar Senaras, M Khalid Khan Niazi, Gerard Lozanski, Metin N Gurcan
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2018-01-01
Series:PLoS ONE
Online Access:http://europepmc.org/articles/PMC6201886?pdf=render
id doaj-f204703c17ca4a7b8798f6462a60a25c
record_format Article
spelling doaj-f204703c17ca4a7b8798f6462a60a25c2020-11-25T00:08:36ZengPublic Library of Science (PLoS)PLoS ONE1932-62032018-01-011310e020538710.1371/journal.pone.0205387DeepFocus: Detection of out-of-focus regions in whole slide digital images using deep learning.Caglar SenarasM Khalid Khan NiaziGerard LozanskiMetin N GurcanThe development of whole slide scanners has revolutionized the field of digital pathology. Unfortunately, whole slide scanners often produce images with out-of-focus/blurry areas that limit the amount of tissue available for a pathologist to make accurate diagnosis/prognosis. Moreover, these artifacts hamper the performance of computerized image analysis systems. These areas are typically identified by visual inspection, which leads to a subjective evaluation causing high intra- and inter-observer variability. Moreover, this process is both tedious, and time-consuming. The aim of this study is to develop a deep learning based software called, DeepFocus, which can automatically detect and segment blurry areas in digital whole slide images to address these problems. DeepFocus is built on TensorFlow, an open source library that exploits data flow graphs for efficient numerical computation. DeepFocus was trained by using 16 different H&E and IHC-stained slides that were systematically scanned on nine different focal planes, generating 216,000 samples with varying amounts of blurriness. When trained and tested on two independent datasets, DeepFocus resulted in an average accuracy of 93.2% (± 9.6%), which is a 23.8% improvement over an existing method. DeepFocus has the potential to be integrated with whole slide scanners to automatically re-scan problematic areas, hence improving the overall image quality for pathologists and image analysis algorithms.http://europepmc.org/articles/PMC6201886?pdf=render
collection DOAJ
language English
format Article
sources DOAJ
author Caglar Senaras
M Khalid Khan Niazi
Gerard Lozanski
Metin N Gurcan
spellingShingle Caglar Senaras
M Khalid Khan Niazi
Gerard Lozanski
Metin N Gurcan
DeepFocus: Detection of out-of-focus regions in whole slide digital images using deep learning.
PLoS ONE
author_facet Caglar Senaras
M Khalid Khan Niazi
Gerard Lozanski
Metin N Gurcan
author_sort Caglar Senaras
title DeepFocus: Detection of out-of-focus regions in whole slide digital images using deep learning.
title_short DeepFocus: Detection of out-of-focus regions in whole slide digital images using deep learning.
title_full DeepFocus: Detection of out-of-focus regions in whole slide digital images using deep learning.
title_fullStr DeepFocus: Detection of out-of-focus regions in whole slide digital images using deep learning.
title_full_unstemmed DeepFocus: Detection of out-of-focus regions in whole slide digital images using deep learning.
title_sort deepfocus: detection of out-of-focus regions in whole slide digital images using deep learning.
publisher Public Library of Science (PLoS)
series PLoS ONE
issn 1932-6203
publishDate 2018-01-01
description The development of whole slide scanners has revolutionized the field of digital pathology. Unfortunately, whole slide scanners often produce images with out-of-focus/blurry areas that limit the amount of tissue available for a pathologist to make accurate diagnosis/prognosis. Moreover, these artifacts hamper the performance of computerized image analysis systems. These areas are typically identified by visual inspection, which leads to a subjective evaluation causing high intra- and inter-observer variability. Moreover, this process is both tedious, and time-consuming. The aim of this study is to develop a deep learning based software called, DeepFocus, which can automatically detect and segment blurry areas in digital whole slide images to address these problems. DeepFocus is built on TensorFlow, an open source library that exploits data flow graphs for efficient numerical computation. DeepFocus was trained by using 16 different H&E and IHC-stained slides that were systematically scanned on nine different focal planes, generating 216,000 samples with varying amounts of blurriness. When trained and tested on two independent datasets, DeepFocus resulted in an average accuracy of 93.2% (± 9.6%), which is a 23.8% improvement over an existing method. DeepFocus has the potential to be integrated with whole slide scanners to automatically re-scan problematic areas, hence improving the overall image quality for pathologists and image analysis algorithms.
url http://europepmc.org/articles/PMC6201886?pdf=render
work_keys_str_mv AT caglarsenaras deepfocusdetectionofoutoffocusregionsinwholeslidedigitalimagesusingdeeplearning
AT mkhalidkhanniazi deepfocusdetectionofoutoffocusregionsinwholeslidedigitalimagesusingdeeplearning
AT gerardlozanski deepfocusdetectionofoutoffocusregionsinwholeslidedigitalimagesusingdeeplearning
AT metinngurcan deepfocusdetectionofoutoffocusregionsinwholeslidedigitalimagesusingdeeplearning
_version_ 1725415541110210560