Journal Article
. 2018 May; 6(3):270-276.
doi: 10.1080/21681163.2016.1141063.

A resolution adaptive deep hierarchical (RADHicaL) learning scheme applied to nuclear segmentation of digital pathology images

Andrew Janowczyk 1 Scott Doyle 2 Hannah Gilmore 3 Anant Madabhushi 1 
Affiliations
  • PMID: 29732269
  •     13 References
  •     10 citations

Abstract

Deep learning (DL) has recently been successfully applied to a number of image analysis problems. However, DL approaches tend to be inefficient for segmentation on large image data, such as high-resolution digital pathology slide images. For example, typical breast biopsy images scanned at 40× magnification contain billions of pixels, of which usually only a small percentage belong to the class of interest. For a typical naïve deep learning scheme, parsing through and interrogating all the image pixels would represent hundreds if not thousands of hours of compute time using high performance computing environments. In this paper, we present a resolution adaptive deep hierarchical (RADHicaL) learning scheme wherein DL networks at lower resolutions are leveraged to determine if higher levels of magnification, and thus computation, are necessary to provide precise results. We evaluate our approach on a nuclear segmentation task with a cohort of 141 ER+ breast cancer images and show we can reduce computation time on average by about 85%. Expert annotations of 12,000 nuclei across these 141 images were employed for quantitative evaluation of RADHicaL. A head-to-head comparison with a naïve DL approach, operating solely at the highest magnification, yielded the following performance metrics: .9407 vs .9854 Detection Rate, .8218 vs .8489 F-score, .8061 vs .8364 true positive rate and .8822 vs 0.8932 positive predictive value. Our performance indices compare favourably with state of the art nuclear segmentation approaches for digital pathology images.

Keywords: Data processing and analysis; applications of imaging and visualisation; deep learning; digital pathology; image processing and analysis; output generation.

Mitosis detection in breast cancer pathology images by combining handcrafted and convolutional neural network features.
Haibo Wang, Angel Cruz-Roa, +6 authors, Anant Madabhushi.
J Med Imaging (Bellingham), 2015 Jul 15; 1(3). PMID: 26158062    Free PMC article.
Comparison of the prognostic value of Scarff-Bloom-Richardson and Nottingham histological grades in a series of 825 cases of breast cancer: major importance of the mitotic count as a component of both grading systems.
C Genestie, B Zafrani, +5 authors, X Sastre-Garau.
Anticancer Res, 1998 May 06; 18(1B). PMID: 9568179
High-throughput biomarker segmentation on ovarian cancer tissue microarrays via hierarchical normalized cuts.
Andrew Janowczyk, Sharat Chandran, +4 authors, Anant Madabhushi.
IEEE Trans Biomed Eng, 2011 Dec 20; 59(5). PMID: 22180503
Multi-field-of-view strategy for image-based outcome prediction of multi-parametric estrogen receptor-positive breast cancer histopathology: Comparison to Oncotype DX.
Ajay Basavanhally, Michael Feldman, +4 authors, Anant Madabhushi.
J Pathol Inform, 2011 Jan 01; 2. PMID: 22811953    Free PMC article.
An integrated region-, boundary-, shape-based active contour for multiple object overlap resolution in histological imagery.
Sahirzeeshan Ali, Anant Madabhushi.
IEEE Trans Med Imaging, 2012 Apr 14; 31(7). PMID: 22498689
A boosted Bayesian multiresolution classifier for prostate cancer detection from digitized needle biopsies.
Scott Doyle, Michael Feldman, John Tomaszewski, Anant Madabhushi.
IEEE Trans Biomed Eng, 2010 Jun 24; 59(5). PMID: 20570758
Gleason grading and prognostic factors in carcinoma of the prostate.
Peter A Humphrey.
Mod Pathol, 2004 Feb 21; 17(3). PMID: 14976540
Review.
Methods for nuclei detection, segmentation, and classification in digital histopathology: a review-current status and future potential.
Humayun Irshad, Antoine Veillard, Ludovic Roux, Daniel Racoceanu.
IEEE Rev Biomed Eng, 2014 May 08; 7. PMID: 24802905
Highly Cited. Review.
Expectation-maximization-driven geodesic active contour with overlap resolution (EMaGACOR): application to lymphocyte segmentation on breast cancer histopathology.
Hussain Fatakdawala, Jun Xu, +5 authors, Anant Madabhushi.
IEEE Trans Biomed Eng, 2010 Feb 23; 57(7). PMID: 20172780
Stacked Sparse Autoencoder (SSAE) for Nuclei Detection on Breast Cancer Histopathology Images.
Jun Xu, Lei Xiang, +4 authors, Anant Madabhushi.
IEEE Trans Med Imaging, 2015 Jul 25; 35(1). PMID: 26208307    Free PMC article.
Highly Cited.
Quantification of histochemical staining by color deconvolution.
A C Ruifrok, D A Johnston.
Anal Quant Cytol Histol, 2001 Sep 04; 23(4). PMID: 11531144
Highly Cited.
Automatic nuclei segmentation in H&E stained breast cancer histopathology images.
Mitko Veta, Paul J van Diest, +3 authors, Josien P W Pluim.
PLoS One, 2013 Aug 08; 8(7). PMID: 23922958    Free PMC article.
Mitosis detection in breast cancer histology images with deep neural networks.
Dan C Cireşan, Alessandro Giusti, Luca M Gambardella, Jürgen Schmidhuber.
Med Image Comput Comput Assist Interv, 2014 Mar 01; 16(Pt 2). PMID: 24579167
Highly Cited.
Advances in the computational and molecular understanding of the prostate cancer cell nucleus.
Neil M Carleton, George Lee, Anant Madabhushi, Robert W Veltri.
J Cell Biochem, 2018 Jun 21; 119(9). PMID: 29923622    Free PMC article.
Review.
Prediction of recurrence in early stage non-small cell lung cancer using computer extracted nuclear features from digital H&E images.
Xiangxue Wang, Andrew Janowczyk, +5 authors, Anant Madabhushi.
Sci Rep, 2017 Oct 21; 7(1). PMID: 29051570    Free PMC article.
Deep learning for digital pathology image analysis: A comprehensive tutorial with selected use cases.
Andrew Janowczyk, Anant Madabhushi.
J Pathol Inform, 2016 Aug 27; 7. PMID: 27563488    Free PMC article.
Highly Cited.
Improving the accuracy of gastrointestinal neuroendocrine tumor grading with deep learning.
Darshana Govind, Kuang-Yu Jen, +6 authors, Pinaki Sarder.
Sci Rep, 2020 Jul 08; 10(1). PMID: 32632119    Free PMC article.
Deep learning-based image analysis methods for brightfield-acquired multiplex immunohistochemistry images.
Danielle J Fassler, Shahira Abousamra, +10 authors, Joel Saltz.
Diagn Pathol, 2020 Jul 30; 15(1). PMID: 32723384    Free PMC article.
The Use of Screencasts with Embedded Whole-Slide Scans and Hyperlinks to Teach Anatomic Pathology in a Supervised Digital Environment.
Mary Wong, Joseph Frye, Stacey Kim, Alberto M Marchevsky.
J Pathol Inform, 2019 Jan 05; 9. PMID: 30607306    Free PMC article.
Convolutional neural network initialized active contour model with adaptive ellipse fitting for nuclear segmentation on breast histopathological images.
Jun Xu, Lei Gong, +4 authors, Anant Madabhushi.
J Med Imaging (Bellingham), 2019 Mar 07; 6(1). PMID: 30840729    Free PMC article.
An integrated iterative annotation technique for easing neural network training in medical image analysis.
Brendon Lutnick, Brandon Ginley, +7 authors, Pinaki Sarder.
Nat Mach Intell, 2019 Jun 13; 1(2). PMID: 31187088    Free PMC article.
Highly Cited.
A prognostic and predictive computational pathology image signature for added benefit of adjuvant chemotherapy in early stage non-small-cell lung cancer.
Xiangxue Wang, Kaustav Bera, +10 authors, Anant Madabhushi.
EBioMedicine, 2021 Jul 16; 69. PMID: 34265509    Free PMC article.
In Situ Classification of Cell Types in Human Kidney Tissue Using 3D Nuclear Staining.
Andre Woloshuk, Suraj Khochare, +10 authors, Seth Winfree.
Cytometry A, 2020 Dec 01; 99(7). PMID: 33252180    Free PMC article.