Deep Learning-Enabled Diagnosis of Liver Adenocarcinoma

Thomas Albrecht; Annik Rossberg; Jana Dorothea Albrecht; Jan Peter Nicolay; Beate Katharina Straub; Tiemo Sven Gerber; Michael Albrecht; Fritz Brinkmann; Alphonse Charbel; Constantin Schwab; Johannes Schreck; Alexander Brobeil; Christa Flechtenmacher; Moritz von Winterfeld; Bruno Christian Köhler; Christoph Springfeld; Arianeb Mehrabi; Stephan Singer; Monika Nadja Vogel; Olaf Neumann; Albrecht Stenzinger; Peter Schirmacher; Cleo-Aron Weis; Stephanie Roessler; Jakob Nikolas Kather; Benjamin Goeppert

doi:10.1053/j.gastro.2023.07.026

Deep Learning-Enabled Diagnosis of Liver Adenocarcinoma

Gastroenterology. 2023 Nov;165(5):1262-1275. doi: 10.1053/j.gastro.2023.07.026. Epub 2023 Aug 9.

Authors

Thomas Albrecht¹, Annik Rossberg², Jana Dorothea Albrecht³, Jan Peter Nicolay³, Beate Katharina Straub⁴, Tiemo Sven Gerber⁴, Michael Albrecht⁵, Fritz Brinkmann², Alphonse Charbel², Constantin Schwab², Johannes Schreck², Alexander Brobeil², Christa Flechtenmacher², Moritz von Winterfeld², Bruno Christian Köhler⁶, Christoph Springfeld⁶, Arianeb Mehrabi⁷, Stephan Singer⁸, Monika Nadja Vogel⁹, Olaf Neumann², Albrecht Stenzinger², Peter Schirmacher¹⁰, Cleo-Aron Weis², Stephanie Roessler¹⁰, Jakob Nikolas Kather¹¹, Benjamin Goeppert¹²

Affiliations

¹ Institute of Pathology, Heidelberg University Hospital, Heidelberg, Germany; Liver Cancer Center Heidelberg, Heidelberg, Germany. Electronic address: Thomas.Albrecht@med.uni-heidelberg.de.
² Institute of Pathology, Heidelberg University Hospital, Heidelberg, Germany.
³ Department of Dermatology, University Medical Centre Mannheim, Mannheim, Germany.
⁴ Institute of Pathology, University Medicine, Johannes Gutenberg University, Mainz, Germany.
⁵ European Center for Angioscience, Medical Faculty of Mannheim, Mannheim, Germany.
⁶ Liver Cancer Center Heidelberg, Heidelberg, Germany; Department of Medical Oncology, National Center for Tumor Diseases, Heidelberg University Hospital, Heidelberg, Germany.
⁷ Liver Cancer Center Heidelberg, Heidelberg, Germany; Department of General, Visceral and Transplantation Surgery, Heidelberg University Hospital, Heidelberg, Germany.
⁸ Institute of Pathology and Neuropathology, Eberhard-Karls University, Tübingen, Germany.
⁹ Diagnostic and Interventional Radiology, Thoraxklinik at Heidelberg University Hospital, Heidelberg, Germany.
¹⁰ Institute of Pathology, Heidelberg University Hospital, Heidelberg, Germany; Liver Cancer Center Heidelberg, Heidelberg, Germany.
¹¹ Department of Medical Oncology, National Center for Tumor Diseases, Heidelberg University Hospital, Heidelberg, Germany; Else Kroener Fresenius Center for Digital Health, Medical Faculty Carl Gustav Carus, Technical University Dresden, Dresden, Germany.
¹² Institute of Pathology, Heidelberg University Hospital, Heidelberg, Germany; Institute of Pathology and Neuropathology, RKH Hospital Ludwigsburg, Ludwigsburg, Germany; Institute of Tissue Medicine and Pathology, University of Bern, Bern, Switzerland.

PMID: 37562657
DOI: 10.1053/j.gastro.2023.07.026

Abstract

Background & aims: Diagnosis of adenocarcinoma in the liver is a frequent scenario in routine pathology and has a critical impact on clinical decision making. However, rendering a correct diagnosis can be challenging, and often requires the integration of clinical, radiologic, and immunohistochemical information. We present a deep learning model (HEPNET) to distinguish intrahepatic cholangiocarcinoma from colorectal liver metastasis, as the most frequent primary and secondary forms of liver adenocarcinoma, with clinical grade accuracy using H&E-stained whole-slide images.

Methods: HEPNET was trained on 714,589 image tiles from 456 patients who were randomly selected in a stratified manner from a pool of 571 patients who underwent surgical resection or biopsy at Heidelberg University Hospital. Model performance was evaluated on a hold-out internal test set comprising 115 patients and externally validated on 159 patients recruited at Mainz University Hospital.

Results: On the hold-out internal test set, HEPNET achieved an area under the receiver operating characteristic curve of 0.994 (95% CI, 0.989-1.000) and an accuracy of 96.522% (95% CI, 94.521%-98.694%) at the patient level. Validation on the external test set yielded an area under the receiver operating characteristic curve of 0.997 (95% CI, 0.995-1.000), corresponding to an accuracy of 98.113% (95% CI, 96.907%-100.000%). HEPNET surpassed the performance of 6 pathology experts with different levels of experience in a reader study of 50 patients (P = .0005), boosted the performance of resident pathologists to the level of senior pathologists, and reduced potential downstream analyses.

Conclusions: We provided a ready-to-use tool with clinical grade performance that may facilitate routine pathology by rendering a definitive diagnosis and guiding ancillary testing. The incorporation of HEPNET into pathology laboratories may optimize the diagnostic workflow, complemented by test-related labor and cost savings.

Keywords: Artificial Intelligence; Biliary Tract Cancer; Digital Pathology; Intestinal Cancer.