Deep learning helps discriminate between autoimmune hepatitis and primary biliary cholangitis
Background & Aims: Biliary abnormalities in autoimmune hepatitis (AIH) and interface hepatitis in primary biliary cholangitis (PBC) occur frequently, and misinterpretation may lead to therapeutic mistakes with a negative impact on patients. This study investigates the use of a deep learning...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-02-01
|
| Series: | JHEP Reports |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2589555924002027 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1825206866533154816 |
|---|---|
| author | Alessio Gerussi Oliver Lester Saldanha Giorgio Cazzaniga Damiano Verda Zunamys I. Carrero Bastian Engel Richard Taubert Francesca Bolis Laura Cristoferi Federica Malinverno Francesca Colapietro Reha Akpinar Luca Di Tommaso Luigi Terracciano Ana Lleo Mauro Viganó Cristina Rigamonti Daniela Cabibi Vincenza Calvaruso Fabio Gibilisco Nicoló Caldonazzi Alessandro Valentino Stefano Ceola Valentina Canini Eugenia Nofit Marco Muselli Julien Calderaro Dina Tiniakos Vincenzo L’Imperio Fabio Pagni Nicola Zucchini Pietro Invernizzi Marco Carbone Jakob Nikolas Kather |
| author_facet | Alessio Gerussi Oliver Lester Saldanha Giorgio Cazzaniga Damiano Verda Zunamys I. Carrero Bastian Engel Richard Taubert Francesca Bolis Laura Cristoferi Federica Malinverno Francesca Colapietro Reha Akpinar Luca Di Tommaso Luigi Terracciano Ana Lleo Mauro Viganó Cristina Rigamonti Daniela Cabibi Vincenza Calvaruso Fabio Gibilisco Nicoló Caldonazzi Alessandro Valentino Stefano Ceola Valentina Canini Eugenia Nofit Marco Muselli Julien Calderaro Dina Tiniakos Vincenzo L’Imperio Fabio Pagni Nicola Zucchini Pietro Invernizzi Marco Carbone Jakob Nikolas Kather |
| author_sort | Alessio Gerussi |
| collection | DOAJ |
| description | Background & Aims: Biliary abnormalities in autoimmune hepatitis (AIH) and interface hepatitis in primary biliary cholangitis (PBC) occur frequently, and misinterpretation may lead to therapeutic mistakes with a negative impact on patients. This study investigates the use of a deep learning (DL)-based pipeline for the diagnosis of AIH and PBC to aid differential diagnosis. Methods: We conducted a multicenter study across six European referral centers, and built a library of digitized liver biopsy slides dating from 1997 to 2023. A training set of 354 cases (266 AIH and 102 PBC) and an external validation set of 92 cases (62 AIH and 30 PBC) were available for analysis. A novel DL model, the autoimmune liver neural estimator (ALNE), was trained on whole-slide images (WSIs) with H&E staining, without human annotations. The ALNE model was evaluated against clinico-pathological diagnoses and tested for interobserver variability among general pathologists. Results: The ALNE model demonstrated high accuracy in differentiating AIH from PBC, achieving an area under the receiver operating characteristic curve of 0.81 in external validation. Attention heatmaps showed that ALNE tends to focus more on areas with increased inflammation, associating such patterns predominantly with AIH. A multivariate explainable ML model revealed that PBC cases misclassified as AIH more often had ALP values between 1 × upper limit of normal (ULN) and 2 × ULN, coupled with AST values above 1 × ULN. Inconsistency among general pathologists was noticed when evaluating a random sample of the same cases (Fleiss’s kappa value 0.09). Conclusions: The ALNE model is the first system generating a quantitative and accurate differential diagnosis between cases with AIH or PBC. Impact and implications: This study demonstrates the significant potential of the autoimmune liver neural estimator model, a transformer-based deep learning system, in accurately distinguishing between autoimmune hepatitis and primary biliary cholangitis using digitized liver biopsy slides without human annotation. The scientific justification for this work lies in addressing the challenge of differentiating these conditions, which often present with overlapping features and can lead to therapeutic mistakes. In addition, there is need for quantitative assessment of information embedded in liver biopsies, which are currently evaluated on qualitative or semi-quantitative methods. The results of this study are crucial for pathologists, researchers, and clinicians, providing a reliable diagnostic tool that reduces interobserver variability and improves diagnostic accuracy of these conditions. Potential methodological limitations, such as the diversity in scanning techniques and slide colorations, were considered, ensuring the robustness and generalizability of the findings. |
| format | Article |
| id | doaj-art-540ee48f8e3e46fcbad534d8c449b20d |
| institution | Kabale University |
| issn | 2589-5559 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Elsevier |
| record_format | Article |
| series | JHEP Reports |
| spelling | doaj-art-540ee48f8e3e46fcbad534d8c449b20d2025-02-07T04:48:06ZengElsevierJHEP Reports2589-55592025-02-0172101198Deep learning helps discriminate between autoimmune hepatitis and primary biliary cholangitisAlessio Gerussi0Oliver Lester Saldanha1Giorgio Cazzaniga2Damiano Verda3Zunamys I. Carrero4Bastian Engel5Richard Taubert6Francesca Bolis7Laura Cristoferi8Federica Malinverno9Francesca Colapietro10Reha Akpinar11Luca Di Tommaso12Luigi Terracciano13Ana Lleo14Mauro Viganó15Cristina Rigamonti16Daniela Cabibi17Vincenza Calvaruso18Fabio Gibilisco19Nicoló Caldonazzi20Alessandro Valentino21Stefano Ceola22Valentina Canini23Eugenia Nofit24Marco Muselli25Julien Calderaro26Dina Tiniakos27Vincenzo L’Imperio28Fabio Pagni29Nicola Zucchini30Pietro Invernizzi31Marco Carbone32Jakob Nikolas Kather33Division of Gastroenterology, Center for Autoimmune Liver Diseases, European Reference Network on Hepatological Diseases (ERN RARE-LIVER), Fondazione IRCCS San Gerardo dei Tintori, Monza, Italy; Department of Medicine and Surgery, University of Milano-Bicocca, Monza, ItalyElse Kroener Fresenius Center for Digital Health, Medical Faculty Carl Gustav Carus, Technical University Dresden, Dresden, Germany; Department of Diagnostic and Interventional Radiology, University Hospital RWTH Aachen, Aachen, GermanyDepartment of Medicine and Surgery, Pathology, Fondazione IRCCS San Gerardo dei Tintori, Università di Milano-Bicocca, Monza, ItalyRulex Innovation Labs, Rulex Inc., Genoa, ItalyElse Kroener Fresenius Center for Digital Health, Medical Faculty Carl Gustav Carus, Technical University Dresden, Dresden, GermanyDepartment of Gastroenterology, Hepatology, Infectious Diseases and Endocrinology, Hannover Medical School, Hannover, Germany; European Reference Network on Hepatological Diseases (ERN RARE-LIVER), Hamburg, GermanyDepartment of Gastroenterology, Hepatology, Infectious Diseases and Endocrinology, Hannover Medical School, Hannover, Germany; European Reference Network on Hepatological Diseases (ERN RARE-LIVER), Hamburg, GermanyDivision of Gastroenterology, Center for Autoimmune Liver Diseases, European Reference Network on Hepatological Diseases (ERN RARE-LIVER), Fondazione IRCCS San Gerardo dei Tintori, Monza, Italy; Department of Medicine and Surgery, University of Milano-Bicocca, Monza, ItalyDivision of Gastroenterology, Center for Autoimmune Liver Diseases, European Reference Network on Hepatological Diseases (ERN RARE-LIVER), Fondazione IRCCS San Gerardo dei Tintori, Monza, Italy; Department of Medicine and Surgery, University of Milano-Bicocca, Monza, ItalyDivision of Gastroenterology, Center for Autoimmune Liver Diseases, European Reference Network on Hepatological Diseases (ERN RARE-LIVER), Fondazione IRCCS San Gerardo dei Tintori, Monza, Italy; Department of Medicine and Surgery, University of Milano-Bicocca, Monza, ItalyDepartment of Biomedical Sciences, Humanitas University, Pieve Emanuele, Milan, Italy; Division of Internal Medicine and Hepatology, Department of Gastroenterology, IRCCS Humanitas Research Hospital, Rozzano, Milan, ItalyDepartment of Biomedical Sciences, Humanitas University, Pieve Emanuele, Milan, Italy; Department of Pathology, IRCSS Humanitas Research Hospital, Rozzano-Milan, ItalyDepartment of Biomedical Sciences, Humanitas University, Pieve Emanuele, Milan, Italy; Department of Pathology, IRCSS Humanitas Research Hospital, Rozzano-Milan, ItalyDepartment of Biomedical Sciences, Humanitas University, Pieve Emanuele, Milan, Italy; Department of Pathology, IRCSS Humanitas Research Hospital, Rozzano-Milan, ItalyDepartment of Biomedical Sciences, Humanitas University, Pieve Emanuele, Milan, Italy; Division of Internal Medicine and Hepatology, Department of Gastroenterology, IRCCS Humanitas Research Hospital, Rozzano, Milan, ItalyGastroenterology Hepatology and Transplantation Unit, ASST Papa Giovanni XXIII, Bergamo, ItalyDepartment of Translational Medicine, Università del Piemonte Orientale, Division of Internal Medicine, AOU Maggiore della Carità, Novara, ItalyPathology Institute, PROMISE, University of Palermo, Palermo, ItalyGastrointestinal and Liver Unit, Department of Health Promotion Sciences, Maternal and Infantile Care, Internal Medicine and Medical Specialties, University of Palermo, Palermo, ItalyDepartment of Pathology, Hospital “Gravina e Santo Pietro”, Caltagirone, Italy; Department of Medical and Surgical Sciences and Advanced Technologies, “G. F. Ingrassia”, University of Catania, Catania, ItalyDepartment of Diagnostics and Public Health, Section of Pathology, University of Verona, Verona, ItalyPathological Unit, Niguarda Hospital, Milan, ItalyDepartment of Medicine and Surgery, Pathology, Fondazione IRCCS San Gerardo dei Tintori, Università di Milano-Bicocca, Monza, ItalyDepartment of Medicine and Surgery, Pathology, Fondazione IRCCS San Gerardo dei Tintori, Università di Milano-Bicocca, Monza, ItalyDivision of Gastroenterology, Center for Autoimmune Liver Diseases, European Reference Network on Hepatological Diseases (ERN RARE-LIVER), Fondazione IRCCS San Gerardo dei Tintori, Monza, Italy; Department of Medicine and Surgery, University of Milano-Bicocca, Monza, ItalyRulex Innovation Labs, Rulex Inc., Genoa, ItalyUniversité Paris Est Créteil, INSERM, IMRB, Créteil, France; Assistance Publique-Hôpitaux de Paris, Henri Mondor-Albert Chenevier University Hospital, Department of Pathology, Créteil, France; Inserm, U955, Team 18, Créteil, FranceDepartment of Pathology, Aretaieion Hospital, Medical School, National and Kapodistrian University of Athens, Athens, Greece; Translational and Clinical Research Institute, Faculty of Medical Sciences, Newcastle University, Newcastle upon Tyne, UKDepartment of Medicine and Surgery, Pathology, Fondazione IRCCS San Gerardo dei Tintori, Università di Milano-Bicocca, Monza, ItalyDepartment of Medicine and Surgery, Pathology, Fondazione IRCCS San Gerardo dei Tintori, Università di Milano-Bicocca, Monza, ItalyDepartment of Medicine and Surgery, Pathology, Fondazione IRCCS San Gerardo dei Tintori, Università di Milano-Bicocca, Monza, ItalyDivision of Gastroenterology, Center for Autoimmune Liver Diseases, European Reference Network on Hepatological Diseases (ERN RARE-LIVER), Fondazione IRCCS San Gerardo dei Tintori, Monza, Italy; Department of Medicine and Surgery, University of Milano-Bicocca, Monza, Italy; Corresponding authors. Addresses: Division of Gastroenterology and Center for Autoimmune Liver Diseases, Department of Medicine and Surgery, University of Milano - Bicocca, Via Cadore 48, 20900 Monza (MB), Italy. Tel.: +39-039-233-2187 (P. Invernizzi); Else Kroener Fresenius Center for Digital Health, Technical University Dresden, Dresden, Germany (J.N. Kather).Department of Medicine and Surgery, University of Milano-Bicocca, Monza, Italy; Liver Unit, ASST Grande Ospedale Metropolitano Niguarda, Milan, ItalyElse Kroener Fresenius Center for Digital Health, Medical Faculty Carl Gustav Carus, Technical University Dresden, Dresden, Germany; Medical Oncology, National Center for Tumor Diseases (NCT), University Hospital Heidelberg, Heidelberg, Germany; Department of Medicine 1, University Hospital and Faculty of Medicine Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany; Corresponding authors. Addresses: Division of Gastroenterology and Center for Autoimmune Liver Diseases, Department of Medicine and Surgery, University of Milano - Bicocca, Via Cadore 48, 20900 Monza (MB), Italy. Tel.: +39-039-233-2187 (P. Invernizzi); Else Kroener Fresenius Center for Digital Health, Technical University Dresden, Dresden, Germany (J.N. Kather).Background & Aims: Biliary abnormalities in autoimmune hepatitis (AIH) and interface hepatitis in primary biliary cholangitis (PBC) occur frequently, and misinterpretation may lead to therapeutic mistakes with a negative impact on patients. This study investigates the use of a deep learning (DL)-based pipeline for the diagnosis of AIH and PBC to aid differential diagnosis. Methods: We conducted a multicenter study across six European referral centers, and built a library of digitized liver biopsy slides dating from 1997 to 2023. A training set of 354 cases (266 AIH and 102 PBC) and an external validation set of 92 cases (62 AIH and 30 PBC) were available for analysis. A novel DL model, the autoimmune liver neural estimator (ALNE), was trained on whole-slide images (WSIs) with H&E staining, without human annotations. The ALNE model was evaluated against clinico-pathological diagnoses and tested for interobserver variability among general pathologists. Results: The ALNE model demonstrated high accuracy in differentiating AIH from PBC, achieving an area under the receiver operating characteristic curve of 0.81 in external validation. Attention heatmaps showed that ALNE tends to focus more on areas with increased inflammation, associating such patterns predominantly with AIH. A multivariate explainable ML model revealed that PBC cases misclassified as AIH more often had ALP values between 1 × upper limit of normal (ULN) and 2 × ULN, coupled with AST values above 1 × ULN. Inconsistency among general pathologists was noticed when evaluating a random sample of the same cases (Fleiss’s kappa value 0.09). Conclusions: The ALNE model is the first system generating a quantitative and accurate differential diagnosis between cases with AIH or PBC. Impact and implications: This study demonstrates the significant potential of the autoimmune liver neural estimator model, a transformer-based deep learning system, in accurately distinguishing between autoimmune hepatitis and primary biliary cholangitis using digitized liver biopsy slides without human annotation. The scientific justification for this work lies in addressing the challenge of differentiating these conditions, which often present with overlapping features and can lead to therapeutic mistakes. In addition, there is need for quantitative assessment of information embedded in liver biopsies, which are currently evaluated on qualitative or semi-quantitative methods. The results of this study are crucial for pathologists, researchers, and clinicians, providing a reliable diagnostic tool that reduces interobserver variability and improves diagnostic accuracy of these conditions. Potential methodological limitations, such as the diversity in scanning techniques and slide colorations, were considered, ensuring the robustness and generalizability of the findings.http://www.sciencedirect.com/science/article/pii/S2589555924002027Computational pathologyDigital pathologyLiverRare liver diseasesAutoimmunityArtificial intelligence |
| spellingShingle | Alessio Gerussi Oliver Lester Saldanha Giorgio Cazzaniga Damiano Verda Zunamys I. Carrero Bastian Engel Richard Taubert Francesca Bolis Laura Cristoferi Federica Malinverno Francesca Colapietro Reha Akpinar Luca Di Tommaso Luigi Terracciano Ana Lleo Mauro Viganó Cristina Rigamonti Daniela Cabibi Vincenza Calvaruso Fabio Gibilisco Nicoló Caldonazzi Alessandro Valentino Stefano Ceola Valentina Canini Eugenia Nofit Marco Muselli Julien Calderaro Dina Tiniakos Vincenzo L’Imperio Fabio Pagni Nicola Zucchini Pietro Invernizzi Marco Carbone Jakob Nikolas Kather Deep learning helps discriminate between autoimmune hepatitis and primary biliary cholangitis JHEP Reports Computational pathology Digital pathology Liver Rare liver diseases Autoimmunity Artificial intelligence |
| title | Deep learning helps discriminate between autoimmune hepatitis and primary biliary cholangitis |
| title_full | Deep learning helps discriminate between autoimmune hepatitis and primary biliary cholangitis |
| title_fullStr | Deep learning helps discriminate between autoimmune hepatitis and primary biliary cholangitis |
| title_full_unstemmed | Deep learning helps discriminate between autoimmune hepatitis and primary biliary cholangitis |
| title_short | Deep learning helps discriminate between autoimmune hepatitis and primary biliary cholangitis |
| title_sort | deep learning helps discriminate between autoimmune hepatitis and primary biliary cholangitis |
| topic | Computational pathology Digital pathology Liver Rare liver diseases Autoimmunity Artificial intelligence |
| url | http://www.sciencedirect.com/science/article/pii/S2589555924002027 |
| work_keys_str_mv | AT alessiogerussi deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT oliverlestersaldanha deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT giorgiocazzaniga deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT damianoverda deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT zunamysicarrero deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT bastianengel deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT richardtaubert deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT francescabolis deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT lauracristoferi deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT federicamalinverno deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT francescacolapietro deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT rehaakpinar deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT lucaditommaso deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT luigiterracciano deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT analleo deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT maurovigano deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT cristinarigamonti deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT danielacabibi deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT vincenzacalvaruso deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT fabiogibilisco deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT nicolocaldonazzi deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT alessandrovalentino deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT stefanoceola deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT valentinacanini deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT eugenianofit deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT marcomuselli deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT juliencalderaro deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT dinatiniakos deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT vincenzolimperio deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT fabiopagni deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT nicolazucchini deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT pietroinvernizzi deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT marcocarbone deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis AT jakobnikolaskather deeplearninghelpsdiscriminatebetweenautoimmunehepatitisandprimarybiliarycholangitis |