Peter Ström1, Kimmo Kartasalo2, Henrik Olsson1, Leslie Solorzano3, Brett Delahunt4, Daniel M Berney5, David G Bostwick6, Andrew J Evans7, David J Grignon8, Peter A Humphrey9, Kenneth A Iczkowski10, James G Kench11, Glen Kristiansen12, Theodorus H van der Kwast7, Katia R M Leite13, Jesse K McKenney14, Jon Oxley15, Chin-Chen Pan16, Hemamali Samaratunga17, John R Srigley18, Hiroyuki Takahashi19, Toyonori Tsuzuki20, Murali Varma21, Ming Zhou22, Johan Lindberg1, Cecilia Lindskog23, Pekka Ruusuvuori2, Carolina Wählby24, Henrik Grönberg25, Mattias Rantalainen1, Lars Egevad26, Martin Eklund27. 1. Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden. 2. Faculty of Medicine and Health Technology, Tampere University, Tampere, Finland. 3. Centre for Image Analysis, Department of Information Technology, Uppsala University, Uppsala, Sweden. 4. Department of Pathology and Molecular Medicine, Wellington School of Medicine and Health Sciences, University of Otago, Wellington, New Zealand. 5. Barts Cancer Institute, Queen Mary University of London, London, UK. 6. Bostwick Laboratories, Orlando, FL, USA. 7. Laboratory Medicine Program, University Health Network, Toronto General Hospital, Toronto, ON, Canada. 8. Department of Pathology and Laboratory Medicine, Indiana University School of Medicine, Indianapolis, IN, USA. 9. Department of Pathology, Yale University School of Medicine, New Haven, CT, USA. 10. Department of Pathology, Medical College of Wisconsin, Milwaukee, WI, USA. 11. Department of Tissue Pathology and Diagnostic Oncology, Royal Prince Alfred Hospital and Central Clinical School, University of Sydney, Sydney, NSW, Australia. 12. Institute of Pathology, University Hospital Bonn, Bonn, Germany. 13. Department of Urology, Laboratory of Medical Research, University of São Paulo Medical School, São Paulo, Brazil. 14. Pathology and Laboratory Medicine Institute, Cleveland Clinic, Cleveland, OH, USA. 15. Department of Cellular Pathology, Southmead Hospital, Bristol, UK. 16. Department of Pathology, Taipei Veterans General Hospital, Taipei, Taiwan. 17. Aquesta Uropathology and University of Queensland, Brisbane, QLD, Australia. 18. Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON, Canada. 19. Department of Pathology, Jikei University School of Medicine, Tokyo, Japan. 20. Department of Surgical Pathology, School of Medicine, Aichi Medical University, Nagakute, Japan. 21. Department of Cellular Pathology, University Hospital of Wales, Cardiff, UK. 22. Department of Pathology, UT Southwestern Medical Center, Dallas, TX, USA. 23. Department of Immunology, Genetics, and Pathology, Uppsala University, Uppsala, Sweden. 24. Centre for Image Analysis, Department of Information Technology, Uppsala University, Uppsala, Sweden; BioImage Informatics Facility of SciLifeLab, Uppsala, Sweden. 25. Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden; Department of Oncology, St Göran Hospital, Stockholm, Sweden. 26. Department of Oncology and Pathology, Karolinska Institutet, Stockholm, Sweden. 27. Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden. Electronic address: martin.eklund@ki.se.
Abstract
BACKGROUND: An increasing volume of prostate biopsies and a worldwide shortage of urological pathologists puts a strain on pathology departments. Additionally, the high intra-observer and inter-observer variability in grading can result in overtreatment and undertreatment of prostate cancer. To alleviate these problems, we aimed to develop an artificial intelligence (AI) system with clinically acceptable accuracy for prostate cancer detection, localisation, and Gleason grading. METHODS: We digitised 6682 slides from needle core biopsies from 976 randomly selected participants aged 50-69 in the Swedish prospective and population-based STHLM3 diagnostic study done between May 28, 2012, and Dec 30, 2014 (ISRCTN84445406), and another 271 from 93 men from outside the study. The resulting images were used to train deep neural networks for assessment of prostate biopsies. The networks were evaluated by predicting the presence, extent, and Gleason grade of malignant tissue for an independent test dataset comprising 1631 biopsies from 246 men from STHLM3 and an external validation dataset of 330 biopsies from 73 men. We also evaluated grading performance on 87 biopsies individually graded by 23 experienced urological pathologists from the International Society of Urological Pathology. We assessed discriminatory performance by receiver operating characteristics and tumour extent predictions by correlating predicted cancer length against measurements by the reporting pathologist. We quantified the concordance between grades assigned by the AI system and the expert urological pathologists using Cohen's kappa. FINDINGS: The AI achieved an area under the receiver operating characteristics curve of 0·997 (95% CI 0·994-0·999) for distinguishing between benign (n=910) and malignant (n=721) biopsy cores on the independent test dataset and 0·986 (0·972-0·996) on the external validation dataset (benign n=108, malignant n=222). The correlation between cancer length predicted by the AI and assigned by the reporting pathologist was 0·96 (95% CI 0·95-0·97) for the independent test dataset and 0·87 (0·84-0·90) for the external validation dataset. For assigning Gleason grades, the AI achieved a mean pairwise kappa of 0·62, which was within the range of the corresponding values for the expert pathologists (0·60-0·73). INTERPRETATION: An AI system can be trained to detect and grade cancer in prostate needle biopsy samples at a ranking comparable to that of international experts in prostate pathology. Clinical application could reduce pathology workload by reducing the assessment of benign biopsies and by automating the task of measuring cancer length in positive biopsy cores. An AI system with expert-level grading performance might contribute a second opinion, aid in standardising grading, and provide pathology expertise in parts of the world where it does not exist. FUNDING: Swedish Research Council, Swedish Cancer Society, Swedish eScience Research Center, EIT Health.
BACKGROUND: An increasing volume of prostate biopsies and a worldwide shortage of urological pathologists puts a strain on pathology departments. Additionally, the high intra-observer and inter-observer variability in grading can result in overtreatment and undertreatment of prostate cancer. To alleviate these problems, we aimed to develop an artificial intelligence (AI) system with clinically acceptable accuracy for prostate cancer detection, localisation, and Gleason grading. METHODS: We digitised 6682 slides from needle core biopsies from 976 randomly selected participants aged 50-69 in the Swedish prospective and population-based STHLM3 diagnostic study done between May 28, 2012, and Dec 30, 2014 (ISRCTN84445406), and another 271 from 93 men from outside the study. The resulting images were used to train deep neural networks for assessment of prostate biopsies. The networks were evaluated by predicting the presence, extent, and Gleason grade of malignant tissue for an independent test dataset comprising 1631 biopsies from 246 men from STHLM3 and an external validation dataset of 330 biopsies from 73 men. We also evaluated grading performance on 87 biopsies individually graded by 23 experienced urological pathologists from the International Society of Urological Pathology. We assessed discriminatory performance by receiver operating characteristics and tumour extent predictions by correlating predicted cancer length against measurements by the reporting pathologist. We quantified the concordance between grades assigned by the AI system and the expert urological pathologists using Cohen's kappa. FINDINGS: The AI achieved an area under the receiver operating characteristics curve of 0·997 (95% CI 0·994-0·999) for distinguishing between benign (n=910) and malignant (n=721) biopsy cores on the independent test dataset and 0·986 (0·972-0·996) on the external validation dataset (benign n=108, malignant n=222). The correlation between cancer length predicted by the AI and assigned by the reporting pathologist was 0·96 (95% CI 0·95-0·97) for the independent test dataset and 0·87 (0·84-0·90) for the external validation dataset. For assigning Gleason grades, the AI achieved a mean pairwise kappa of 0·62, which was within the range of the corresponding values for the expert pathologists (0·60-0·73). INTERPRETATION: An AI system can be trained to detect and grade cancer in prostate needle biopsy samples at a ranking comparable to that of international experts in prostate pathology. Clinical application could reduce pathology workload by reducing the assessment of benign biopsies and by automating the task of measuring cancer length in positive biopsy cores. An AI system with expert-level grading performance might contribute a second opinion, aid in standardising grading, and provide pathology expertise in parts of the world where it does not exist. FUNDING: Swedish Research Council, Swedish Cancer Society, Swedish eScience Research Center, EIT Health.
Authors: Andreas Kleppe; Ole-Johan Skrede; Sepp De Raedt; Knut Liestøl; David J Kerr; Håvard E Danielsen Journal: Nat Rev Cancer Date: 2021-01-29 Impact factor: 60.716
Authors: Oscar E Brück; Susanna E Lallukka-Brück; Helena R Hohtari; Aleksandr Ianevski; Freja T Ebeling; Panu E Kovanen; Soili I Kytölä; Tero A Aittokallio; Pedro M Ramos; Kimmo V Porkka; Satu M Mustjoki Journal: Blood Cancer Discov Date: 2021-03-22
Authors: N Harland; B Amend; N Lipke; S Y Brucker; F Fend; A Herkommer; H Lensch; O Sawodny; T E Schäffer; K Schenke-Layland; C Tarín Sauer; W Aicher; A Stenzl Journal: Urologe A Date: 2021-07-13 Impact factor: 0.639
Authors: Drew F K Williamson; Tiffany Y Chen; Ming Y Lu; Richard J Chen; Matteo Barbieri; Faisal Mahmood Journal: Nat Biomed Eng Date: 2021-03-01 Impact factor: 25.671