Epithelium segmentation and automated Gleason grading of prostate cancer via deep learning in label‐free multiphoton microscopic images

In the current clinical care practice, Gleason grading system is one of the most powerful prognostic predictors for prostate cancer (PCa). The grading system is based on the architectural pattern of cancerous epithelium in histological images. However, the standard procedure of histological examination often involves complicated tissue fixation and staining, which are time-consuming and may delay the diagnosis and surgery. In this study, label-free multiphoton microscopy (MPM) was used to acquire subcellular-resolution images of unstained prostate tissues. Then, a deep learning architecture (U-net) was introduced for epithelium segmentation of prostate tissues in MPM images. The obtained segmentation results were then merged with the original MPM images to train a classification network (AlexNet) for automated Gleason grading. The developed method achieved an overall pixel accuracy of 92.3% with a mean F1 score of 0.839 for epithelium segmentation. By merging the segmentation results with the MPM images, the accuracy of Gleason grading was improved from 72.42% to 81.13% in hold-out test set. Our results suggest that MPM in combination with deep learning holds the potential to be used as a fast and powerful clinical tool for PCa diagnosis. This article is protected by copyright. All rights reserved.

[1]  J. Epstein.  An update of the Gleason grading system. , 2010, The Journal of urology.

[2]  Paul Campagnola,et al.  Second harmonic generation imaging microscopy: applications to diseases diagnostics. , 2011, Analytical chemistry.

[3]  Stephen A. Boppart,et al.  Stain-free histopathology by programmable supercontinuum pulses , 2016, Nature Photonics.

[4]  Yuting Ling,et al.  Second harmonic generation (SHG) imaging of cancer heterogeneity in ultrasound guided biopsies of prostate in men suspected with prostate cancer , 2017, Journal of biophotonics.

[5]  Antonio Lopez-Beltran,et al.  Current practice of Gleason grading of prostate carcinoma , 2006, Virchows Archiv.

[6]  W. Webb,et al.  Live tissue intrinsic emission microscopy using multiphoton-excited native fluorescence and second harmonic generation , 2003, Proceedings of the National Academy of Sciences of the United States of America.

[7]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[8]  Woei-Chyn Chu,et al.  Performance measure characterization for evaluating neuroimage segmentation algorithms , 2009, NeuroImage.

[9]  B. Delahunt,et al.  The 2014 International Society of Urological Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic Carcinoma: Definition of Grading Patterns and Proposal for a New Grading System , 2015, The American journal of surgical pathology.

[10]  Francesco Saverio Pavone,et al.  In vivo non‐invasive monitoring of collagen remodelling by two‐photon microscopy after micro‐ablative fractional laser resurfacing , 2014, Journal of biophotonics.

[11]  Yuanxiang Lin,et al.  Label-free detection of brain invasion in meningiomas by multiphoton microscopy , 2018, Laser Physics Letters.

[12]  Ronald M. Summers,et al.  Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning , 2016, IEEE Transactions on Medical Imaging.

[13]  Hao Chen,et al.  MILD‐Net: Minimal information loss dilated network for gland instance segmentation in colon histology images , 2018, Medical Image Anal..

[14]  Robert W Boyd,et al.  Automated classification of multiphoton microscopy images of ovarian tissue using deep learning , 2018, Journal of biomedical optics.

[15]  B. Tromberg,et al.  Imaging cells and extracellular matrix in vivo by using second-harmonic generation and two-photon excited fluorescence , 2002, Proceedings of the National Academy of Sciences of the United States of America.

[16]  Tayyar A. Ozkan,et al.  Interobserver variability in Gleason histological grading of prostate cancer , 2016, Scandinavian journal of urology.

[17]  Christophe Odin,et al.  Fibrillar collagen scoring by second harmonic microscopy: a new tool in the assessment of liver fibrosis. , 2010, Journal of hepatology.

[18]  Shuangmu Zhuo,et al.  Multimode nonlinear optical imaging of the dermis in ex vivo human skin based on the combination of multichannel mode and Lambda mode. , 2006, Optics express.

[19]  T. Johnson,et al.  Detection of human brain tumor infiltration with quantitative stimulated Raman scattering microscopy , 2015, Science Translational Medicine.

[20]  Minsue T. Chen,et al.  Hematoxylin and Eosin Tissue Stain in Mohs Micrographic Surgery: A Review , 2011, Dermatologic surgery : official publication for American Society for Dermatologic Surgery [et al.].

[21]  Urs Utzinger,et al.  Endogenous Optical Biomarkers of Ovarian Cancer Evaluated with Multiphoton Microscopy , 2007, Cancer Epidemiology Biomarkers & Prevention.

[22]  William R Wagner,et al.  A custom image-based analysis tool for quantifying elastin and collagen micro-architecture in the wall of the human aorta from multi-photon microscopy. , 2014, Journal of biomechanics.

[23]  Nima Tajbakhsh,et al.  Convolutional Neural Networks for Medical Image Analysis: Full Training or Fine Tuning? , 2016, IEEE Transactions on Medical Imaging.

[24]  Oleg Nadiarnykh,et al.  Alterations of the extracellular matrix in ovarian cancer studied by Second Harmonic Generation imaging microscopy , 2010, BMC Cancer.

[25]  Hao Chen,et al.  DCAN: Deep contour‐aware networks for object instance segmentation from histology images , 2017, Medical Image Anal..

[26]  Gang Wang,et al.  Automatic grading of prostate cancer in digitized histopathology images: Learning from multiple experts , 2018, Medical Image Anal..

[27]  David B. A. Epstein,et al.  Micro‐Net: A unified model for segmentation of various objects in microscopy images , 2018, Medical Image Anal..

[28]  A. Jemal,et al.  Cancer statistics, 2017 , 2017, CA: a cancer journal for clinicians.

[29]  Andrew Janowczyk,et al.  Deep learning for digital pathology image analysis: A comprehensive tutorial with selected use cases , 2016, Journal of pathology informatics.

[30]  T. Hermanns,et al.  Automated Gleason grading of prostate cancer tissue microarrays via deep learning , 2018, Scientific Reports.

[31]  M. Varma,et al.  Diagnostic utility of immunohistochemistry in morphologically difficult prostate cancer: review of current literature , 2005, Histopathology.

[32]  Kimani C Toussaint,et al.  Third‐harmonic generation imaging of breast tissue biopsies , 2016, Journal of microscopy.

[33]  W. Zipfel,et al.  Strategies for high-resolution imaging of epithelial ovarian cancer by laparoscopic nonlinear microscopy. , 2010, Translational oncology.

[34]  B. van Ginneken,et al.  Deep learning as a tool for increased accuracy and efficiency of histopathological diagnosis , 2016, Scientific Reports.

[35]  Abhishek Srivastava,et al.  Multiphoton microscopy for structure identification in human prostate and periprostatic tissue: implications in prostate cancer surgery , 2011, BJU international.

[36]  Bram van Ginneken,et al.  A survey on deep learning in medical image analysis , 2017, Medical Image Anal..