Fully automated quantitative assessment of hepatic steatosis in liver transplants

BACKGROUND The presence of macro- and microvesicular steatosis is one of the major risk factors for liver transplantation. An accurate assessment of the steatosis percentage is crucial for determining liver graft transplantability, which is currently based on the pathologists' visual evaluations on liver histology specimens. METHOD The aim of this study was to develop and validate a fully automated algorithm, called HEPASS (HEPatic Adaptive Steatosis Segmentation), for both micro- and macro-steatosis detection in digital liver histological images. The proposed method employs a hybrid deep learning framework, combining the accuracy of an adaptive threshold with the semantic segmentation of a deep convolutional neural network. Starting from all white regions, the HEPASS algorithm was able to detect lipid droplets and classify them into micro- or macrosteatosis. RESULTS The proposed method was developed and tested on 385 hematoxylin and eosin (H&E) stained images coming from 77 liver donors. Automated results were compared with manual annotations and nine state-of-the-art techniques designed for steatosis segmentation. In the TEST set, the algorithm was characterized by 97.27% accuracy in steatosis quantification (average error 1.07%, maximum average error 5.62%) and outperformed all the compared methods. CONCLUSIONS To the best of our knowledge, the proposed algorithm is the first fully automated algorithm for the assessment of both micro- and macrosteatosis in H&E stained liver tissue images. Being very fast (average computational time 0.72 s), this algorithm paves the way for automated, quantitative and real-time liver graft assessments.

[1]  Tony F. Chan,et al.  Active contours without edges , 2001, IEEE Trans. Image Process..

[2]  Bruno Turlin,et al.  Assessment of hepatic steatosis: comparison of quantitative and semiquantitative methods in 108 liver biopsies , 2009, Liver international : official journal of the International Association for the Study of the Liver.

[3]  Victor Hugo C. de Albuquerque,et al.  Novel Adaptive Balloon Active Contour Method based on internal force for image segmentation - A systematic evaluation on synthetic and real images , 2014, Expert Syst. Appl..

[4]  C. Lo,et al.  Excellent outcomes of liver transplantation using severely steatotic grafts from brain‐dead donors , 2016, Liver transplantation : official publication of the American Association for the Study of Liver Diseases and the International Liver Transplantation Society.

[5]  Alessandro Ricchiuti,et al.  Marginal grafts: finding the correct treatment for fatty livers , 2003, Transplant international : official journal of the European Society for Organ Transplantation.

[6]  Andrew R. Hall,et al.  Hepatic steatosis estimated microscopically versus digital image analysis , 2013, Liver international : official journal of the International Association for the Study of the Liver.

[7]  J. A. van der Laak,et al.  A Novel Automatic Digital Algorithm that Accurately Quantifies Steatosis in NAFLD on Histopathological Whole‐Slide Images , 2019, Cytometry. Part B, Clinical cytometry.

[8]  Dimitris N. Metaxas,et al.  tlas-based liver segmentation and hepatic fat-fraction assessment or clinical trials , 2014 .

[9]  Mrinal K. Mandal,et al.  Automated analysis and classification of melanocytic tumor on skin whole slide images , 2018, Comput. Medical Imaging Graph..

[10]  K. Jauch,et al.  Effect of graft steatosis on liver function and organ survival after liver transplantation. , 2008, American journal of surgery.

[11]  D. Mirza,et al.  The marginal liver donor – an update , 2008, Transplant international : official journal of the European Society for Organ Transplantation.

[12]  R. Busuttil,et al.  The utility of marginal donors in liver transplantation , 2003, Liver transplantation : official publication of the American Association for the Study of Liver Diseases and the International Liver Transplantation Society.

[13]  E. Brunt.  Nonalcoholic Steatohepatitis: Definition and Pathology , 2001, Seminars in liver disease.

[14]  Filippo Molinari,et al.  Automatic discrimination of neoplastic epithelium and stromal response in breast carcinoma , 2019, Comput. Biol. Medicine.

[15]  Alexandros T. Tzallas,et al.  Automated Detection of Liver Histopathological Findings Based on Biopsy Image Processing , 2017, Inf..

[16]  N. Chalasani,et al.  Expanding the Donor Pool With the Use of Extended Criteria Donation After Circulatory Death Livers , 2019, Liver transplantation : official publication of the American Association for the Study of Liver Diseases and the International Liver Transplantation Society.

[17]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[18]  Philipp Dutkowski,et al.  Liver transplantation using fatty livers: always feasible? , 2011, Journal of hepatology.

[19]  M. Stegall,et al.  Risk factors for primary dysfunction after liver transplantation--a multivariate analysis. , 1993, Transplantation.

[20]  O. Cummings,et al.  Design and validation of a histological scoring system for nonalcoholic fatty liver disease , 2005, Hepatology.

[21]  K. Washington,et al.  Assessment of Hepatic Steatosis by Expert Pathologists: The End of a Gold Standard , 2009, Annals of surgery.

[22]  M. Deschênes.  Early allograft dysfunction: Causes, recognition, and management , 2013, Liver transplantation : official publication of the American Association for the Study of Liver Diseases and the International Liver Transplantation Society.

[23]  J. Trotter.  Liver transplantation around the world , 2017, Current opinion in organ transplantation.

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

[25]  S. Sherlock,et al.  Diseases of the Liver and Biliary System , 1955 .

[26]  Francois Berthiaume,et al.  Automated image analysis method for detecting and quantifying macrovesicular steatosis in hematoxylin and eosin–stained histology images of human livers , 2014, Liver transplantation : official publication of the American Association for the Study of Liver Diseases and the International Liver Transplantation Society.

[27]  N. Otsu.  A threshold selection method from gray level histograms , 1979 .

[28]  S. Saab,et al.  Major Challenges Limiting Liver Transplantation in the United States , 2011, American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons.

[29]  K. Sugimachi,et al.  Use of steatotic graft in living-donor liver transplantation , 2003, Transplantation.

[30]  C. Mandarim-de-Lacerda,et al.  A critical analysis of three quantitative methods of assessment of hepatic steatosis in liver biopsies , 2011, Virchows Archiv.

[31]  Metin Nafi Gürcan,et al.  Heterogeneity assessment of histological tissue sections in whole slide images , 2015, Comput. Medical Imaging Graph..

[32]  J. Kench,et al.  Grade of deceased donor liver macrovesicular steatosis impacts graft and recipient outcomes more than the Donor Risk Index , 2012, Journal of gastroenterology and hepatology.

[33]  Joseph Bockhorst,et al.  Automatic classification of white regions in liver biopsies by supervised machine learning. , 2014, Human pathology.

[34]  Nazre Batool.  Detection and spatial analysis of hepatic steatosis in histopathology images using sparse linear models , 2016, 2016 Sixth International Conference on Image Processing Theory, Tools and Applications (IPTA).

[35]  Pierre-Alain Clavien,et al.  When is steatosis too much for transplantation? , 2006, Journal of hepatology.

[36]  D Cascella,et al.  An innovative methodology for the automated morphometric and quantitative estimation of liver steatosis. , 2009, Histology and histopathology.

[37]  Reiner Eschbach,et al.  Spatial Color-to-Grayscale Transform Preserving Chrominance Edge Information , 2004, CIC.

[38]  B. Kasiske,et al.  Value of liver biopsy in the evaluation and management of chronic liver disease in renal transplant recipients. , 1993, The American journal of medicine.

[39]  C. Kin,et al.  Extended-donor criteria liver allografts. , 2006, Seminars in liver disease.

[40]  Filippo Molinari,et al.  Multi-tissue and multi-scale approach for nuclei segmentation in H&E stained images , 2018, Biomedical engineering online.

[41]  Filippo Molinari,et al.  Stain Color Adaptive Normalization (SCAN) algorithm: Separation and standardization of histological stains in digital pathology , 2020, Comput. Methods Programs Biomed..

[42]  N. Giannakeas,et al.  Fat Quantitation in Liver Biopsies Using a Pretrained Classification Based System , 2018, Engineering, Technology & Applied Science Research.