30
views
0
recommends
+1 Recommend
0 collections
    0
    shares
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      Epithelium segmentation using deep learning in H&E-stained prostate specimens with immunohistochemistry as reference standard

      research-article

      Read this article at

      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          Given the importance of gland morphology in grading prostate cancer (PCa), automatically differentiating between epithelium and other tissues is an important prerequisite for the development of automated methods for detecting PCa. We propose a new deep learning method to segment epithelial tissue in digitised hematoxylin and eosin (H&E) stained prostatectomy slides using immunohistochemistry (IHC) as reference standard. We used IHC to create a precise and objective ground truth compared to manual outlining on H&E slides, especially in areas with high-grade PCa. 102 tissue sections were stained with H&E and subsequently restained with P63 and CK8/18 IHC markers to highlight epithelial structures. Afterwards each pair was co-registered. First, we trained a U-Net to segment epithelial structures in IHC using a subset of the IHC slides that were preprocessed with color deconvolution. Second, this network was applied to the remaining slides to create the reference standard used to train a second U-Net on H&E. Our system accurately segmented both intact glands and individual tumour epithelial cells. The generalisation capacity of our system is shown using an independent external dataset from a different centre. We envision this segmentation as the first part of a fully automated prostate cancer grading pipeline.

          Related collections

          Most cited references9

          • Record: found
          • Abstract: found
          • Article: found
          Is Open Access

          Deep learning as a tool for increased accuracy and efficiency of histopathological diagnosis

          Pathologists face a substantial increase in workload and complexity of histopathologic cancer diagnosis due to the advent of personalized medicine. Therefore, diagnostic protocols have to focus equally on efficiency and accuracy. In this paper we introduce ‘deep learning’ as a technique to improve the objectivity and efficiency of histopathologic slide analysis. Through two examples, prostate cancer identification in biopsy specimens and breast cancer metastasis detection in sentinel lymph nodes, we show the potential of this new methodology to reduce the workload for pathologists, while at the same time increasing objectivity of diagnoses. We found that all slides containing prostate cancer and micro- and macro-metastases of breast cancer could be identified automatically while 30–40% of the slides containing benign and normal tissue could be excluded without the use of any additional immunohistochemical markers or human intervention. We conclude that ‘deep learning’ holds great promise to improve the efficacy of prostate cancer diagnosis and breast cancer staging.
            Bookmark
            • Record: found
            • Abstract: found
            • Article: not found

            An update of the Gleason grading system.

            An update is provided of the Gleason grading system, which has evolved significantly since its initial description. A search was performed using the MEDLINE(R) database and referenced lists of relevant studies to obtain articles concerning changes to the Gleason grading system. Since the introduction of the Gleason grading system more than 40 years ago many aspects of prostate cancer have changed, including prostate specific antigen testing, transrectal ultrasound guided prostate needle biopsy with greater sampling, immunohistochemistry for basal cells that changed the classification of prostate cancer and new prostate cancer variants. The system was updated at a 2005 consensus conference of international experts in urological pathology, under the auspices of the International Society of Urological Pathology. Gleason score 2-4 should rarely if ever be diagnosed on needle biopsy, certain patterns (ie poorly formed glands) originally considered Gleason pattern 3 are now considered Gleason pattern 4 and all cribriform cancer should be graded pattern 4. The grading of variants and subtypes of acinar adenocarcinoma of the prostate, including cancer with vacuoles, foamy gland carcinoma, ductal adenocarcinoma, pseudohyperplastic carcinoma and small cell carcinoma have also been modified. Other recent issues include reporting secondary patterns of lower and higher grades when present to a limited extent, and commenting on tertiary grade patterns which differ depending on whether the specimen is from needle biopsy or radical prostatectomy. Whereas there is little debate on the definition of tertiary pattern on needle biopsy, this issue is controversial in radical prostatectomy specimens. Although tertiary Gleason patterns are typically added to pathology reports, they are routinely omitted in practice since there is no simple way to incorporate them in predictive nomograms/tables, research studies and patient counseling. Thus, a modified radical prostatectomy Gleason scoring system was recently proposed to incorporate tertiary Gleason patterns in an intuitive fashion. For needle biopsy with different cores showing different grades, the current recommendation is to report the grades of each core separately, whereby the highest grade tumor is selected as the grade of the entire case to determine treatment, regardless of the percent involvement. After the 2005 consensus conference several studies confirmed the superiority of the modified Gleason system as well as its impact on urological practice. It is remarkable that nearly 40 years after its inception the Gleason grading system remains one of the most powerful prognostic factors for prostate cancer. This system has remained timely because of gradual adaptations by urological pathologists to accommodate the changing practice of medicine. Copyright 2010 American Urological Association. Published by Elsevier Inc. All rights reserved.
              Bookmark
              • Record: found
              • Abstract: found
              • Article: not found

              Machine learning approaches to analyze histological images of tissues from radical prostatectomies

              Computerized evaluation of histological preparations of prostate tissues involves identification of tissue components such as stroma (ST), benign/normal epithelium (BN) and prostate cancer (PCa). Image classification approaches have been developed to identify and classify glandular regions in digital images of prostate tissues; however their success has been limited by difficulties in cellular segmentation and tissue heterogeneity. We hypothesized that utilizing image pixels to generate intensity histograms of hematoxylin (H) and eosin (E) stains deconvoluted from H&E images numerically captures the architectural difference between glands and stroma. In addition, we postulated that joint histograms of local binary patterns and local variance (LBPxVAR) can be used as sensitive textural features to differentiate benign/normal tissue from cancer. Here we utilized a machine learning approach comprising of a support vector machine (SVM) followed by a random forest (RF) classifier to digitally stratify prostate tissue into ST, BN and PCa areas. Two pathologists manually annotated 210 images of low- and high-grade tumors from slides that were selected from 20 radical prostatectomies and digitized at high-resolution. The 210 images were split into the training ( n = 19) and test ( n = 191) sets. Local intensity histograms of H and E were used to train a SVM classifier to separate ST from epithelium (BN + PCa). The performance of SVM prediction was evaluated by measuring the accuracy of delineating epithelial areas. The Jaccard J = 59.5 ± 14.6 and Rand Ri = 62.0 ± 7.5 indices reported a significantly better prediction when compared to a reference method (Chen et al., Clinical Proteomics 2013, 10:18) based on the averaged values from the test set. To distinguish BN from PCa we trained a RF classifier with LBPxVAR and local intensity histograms and obtained separate performance values for BN and PCa: J BN = 35.2 ± 24.9, O BN = 49.6 ± 32, J PCa = 49.5 ± 18.5, O PCa = 72.7 ± 14.8 and Ri = 60.6 ± 7.6 in the test set. Our pixel-based classification does not rely on the detection of lumens, which is prone to errors and has limitations in high-grade cancers and has the potential to aid in clinical studies in which the quantification of tumor content is necessary to prognosticate the course of the disease. The image data set with ground truth annotation is available for public use to stimulate further research in this area.
                Bookmark

                Author and article information

                Contributors
                wouter.bulten@radboudumc.nl
                Journal
                Sci Rep
                Sci Rep
                Scientific Reports
                Nature Publishing Group UK (London )
                2045-2322
                29 January 2019
                29 January 2019
                2019
                : 9
                : 864
                Affiliations
                [1 ]ISNI 0000 0004 0444 9382, GRID grid.10417.33, Radboud University Medical Center, Diagnostic Image Analysis Group and the Department of Pathology, ; 6500HB Nijmegen, The Netherlands
                [2 ]ISNI 0000 0004 0444 9382, GRID grid.10417.33, Radboud University Medical Center, Department of Pathology, ; 6500HB Nijmegen, The Netherlands
                [3 ]ISNI 0000 0004 0496 8246, GRID grid.428590.2, Fraunhofer MEVIS, ; 23562 Lübeck, Germany
                [4 ]ISNI 0000 0004 0444 9382, GRID grid.10417.33, Radboud University Medical Center, Diagnostic Image Analysis Group and the Department of Radiology and Nuclear Medicine, ; 6500HB Nijmegen, The Netherlands
                Author information
                http://orcid.org/0000-0002-6129-5039
                http://orcid.org/0000-0003-3387-2596
                http://orcid.org/0000-0001-7982-0754
                http://orcid.org/0000-0003-1554-1291
                Article
                37257
                10.1038/s41598-018-37257-4
                6351532
                30696866
                ffca8600-01a2-49a4-8abe-073d0457c2db
                © The Author(s) 2019

                Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

                History
                : 1 October 2018
                : 3 December 2018
                Categories
                Article
                Custom metadata
                © The Author(s) 2019

                Uncategorized
                Uncategorized

                Comments

                Comment on this article