1
|
Pulaski H, Mehta SS, Manigat LC, Kaufman S, Hou H, Nalbantoglu ILK, Zhang X, Curl E, Taliano R, Kim TH, Torbenson M, Glickman JN, Resnick MB, Patel N, Taylor CE, Bedossa P, Montalto MC, Beck AH, Wack KE. Validation of a whole slide image management system for metabolic-associated steatohepatitis for clinical trials. J Pathol Clin Res 2024; 10:e12395. [PMID: 39294925 PMCID: PMC11410674 DOI: 10.1002/2056-4538.12395] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/01/2023] [Revised: 07/04/2024] [Accepted: 07/15/2024] [Indexed: 09/21/2024]
Abstract
The gold standard for enrollment and endpoint assessment in metabolic dysfunction-associated steatosis clinical trials is histologic assessment of a liver biopsy performed on glass slides. However, obtaining the evaluations from several expert pathologists on glass is challenging, as shipping the slides around the country or around the world is time-consuming and comes with the hazards of slide breakage. This study demonstrated that pathologic assessment of disease activity in steatohepatitis, performed using digital images on the AISight whole slide image management system, yields results that are comparable to those obtained using glass slides. The accuracy of scoring for steatohepatitis (nonalcoholic fatty liver disease activity score ≥4 with ≥1 for each feature and absence of atypical features suggestive of other liver disease) performed on the system was evaluated against scoring conducted on glass slides. Both methods were assessed for overall percent agreement with a consensus "ground truth" score (defined as the median score of a panel of three pathologists' glass slides). Each case was also read by three different pathologists, once on glass and once digitally with a minimum 2-week washout period between the modalities. It was demonstrated that the average agreement across three pathologists of digital scoring with ground truth was noninferior to the average agreement of glass scoring with ground truth [noninferiority margin: -0.05; difference: -0.001; 95% CI: (-0.027, 0.026); and p < 0.0001]. For each pathologist, there was a similar average agreement of digital and glass reads with glass ground truth (pathologist A, 0.843 and 0.849; pathologist B, 0.633 and 0.605; and pathologist C, 0.755 and 0.780). Here, we demonstrate that the accuracy of digital reads for steatohepatitis using digital images is equivalent to glass reads in the context of a clinical trial for scoring using the Clinical Research Network scoring system.
Collapse
Affiliation(s)
| | | | | | | | | | - ILKe Nalbantoglu
- Department of Pathology, Yale University School of Medicine, New Haven, CT, USA
| | - Xuchen Zhang
- Department of Pathology, Yale University School of Medicine, New Haven, CT, USA
| | - Emily Curl
- Foundation Medicine, Inc, Boston, MA, USA
| | - Ross Taliano
- Department of Pathology, Rhode Island Hospital, Providence, RI, USA
| | - Tae Hun Kim
- Pathology Medical Group of Riverside, Providence, RI, USA
| | - Michael Torbenson
- Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, MN, USA
| | | | | | | | | | | | | | | | | |
Collapse
|
2
|
van den Brand M, Nooijen PTGA, van der Laan KD, de Bruin PC, van Leeuwen AMG, Leeuwis JW, Meijer JW, Otte-Höller I, Hebeda KM. Discrepancies in digital hematopathology diagnoses for consultation and expert panel analysis. Virchows Arch 2021; 478:535-540. [PMID: 32840673 PMCID: PMC7973407 DOI: 10.1007/s00428-020-02907-4] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/06/2019] [Revised: 07/23/2020] [Accepted: 08/10/2020] [Indexed: 10/26/2022]
Abstract
Digital pathology with whole-slide imaging (WSI) has a large potential to make the process of expert consultation and expert panel diagnosis more rapid and more efficient. However, comparison with the current methods is necessary for validation of the technique. In this study, we determined if digital assessment of whole-slide images of hematopathology specimens with a focus on the assessment of lymphoma can be used for consultation and panel diagnostics. Ninety-three histological specimens with a suspicion for lymphoma were assessed both with conventional microscopy and digital microscopy with a wash out period between assessments. A consensus diagnosis was based on full concordance between the pathologists or, in case of discordances, was reached at a joint session at a multi-headed microscope. In 81% of the cases, there was a full concordance between digital and light microscopical assessment for all three pathologists. Discordances between conventional microscopy and digital pathology were present in 3% of assessments. In comparison with the consensus diagnosis, discordant diagnoses were made in 5 cases with digital microscopy and in 3 cases with light microscopy. The reported level of confidence and need for additional investigations were similar between assessment by conventional and by digital microscopy. In conclusion, the performance of assessment by digital pathology is in general comparable with that of conventional light microscopy and pathologists feel confident using digital pathology for this subspecialty.
Collapse
Affiliation(s)
- Michiel van den Brand
- Department of Pathology, Radboud University Medical Center, P. O. Box 9101, 6500, HB, Nijmegen, The Netherlands.
- Pathology-DNA, Rijnstate Hospital, Arnhem, The Netherlands.
| | | | - Kimberly D van der Laan
- Department of Pathology, Radboud University Medical Center, P. O. Box 9101, 6500, HB, Nijmegen, The Netherlands
| | - Peter C de Bruin
- Pathology-DNA, St. Antonius Hospital, Nieuwegein, The Netherlands
| | | | | | - Jos W Meijer
- Pathology-DNA, Rijnstate Hospital, Arnhem, The Netherlands
| | - Irene Otte-Höller
- Department of Pathology, Radboud University Medical Center, P. O. Box 9101, 6500, HB, Nijmegen, The Netherlands
| | - Konnie M Hebeda
- Department of Pathology, Radboud University Medical Center, P. O. Box 9101, 6500, HB, Nijmegen, The Netherlands
| |
Collapse
|
3
|
Chong Y, Kim DC, Jung CK, Kim DC, Song SY, Joo HJ, Yi SY. Recommendations for pathologic practice using digital pathology: consensus report of the Korean Society of Pathologists. J Pathol Transl Med 2020; 54:437-452. [PMID: 33027850 PMCID: PMC7674756 DOI: 10.4132/jptm.2020.08.27] [Citation(s) in RCA: 14] [Impact Index Per Article: 3.5] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/14/2020] [Accepted: 08/27/2020] [Indexed: 11/17/2022] Open
Abstract
Digital pathology (DP) using whole slide imaging (WSI) is becoming a fundamental issue in pathology with recent advances and the rapid development of associated technologies. However, the available evidence on its diagnostic uses and practical advice for pathologists on implementing DP remains insufficient, particularly in light of the exponential growth of this industry. To inform DP implementation in Korea, we developed relevant and timely recommendations. We first performed a literature review of DP guidelines, recommendations, and position papers from major countries, as well as a review of relevant studies validating WSI. Based on that information, we prepared a draft. After several revisions, we released this draft to the public and the members of the Korean Society of Pathologists through our homepage and held an open forum for interested parties. Through that process, this final manuscript has been prepared. This recommendation contains an overview describing the background, objectives, scope of application, and basic terminology; guidelines and considerations for the hardware and software used in DP systems and the validation required for DP implementation; conclusions; and references and appendices, including literature on DP from major countries and WSI validation studies.
Collapse
Affiliation(s)
- Yosep Chong
- Department of Hospital Pathology, College of Medicine, The Catholic University of Korea, Seoul, Korea
| | - Dae Cheol Kim
- Department of Pathology, Dong-A University College of Medicine, Busan, Korea
| | - Chan Kwon Jung
- Department of Hospital Pathology, College of Medicine, The Catholic University of Korea, Seoul, Korea
| | - Dong-chul Kim
- Department of Pathology, Seoul Clinical Laboratories, Yongin, Korea
| | - Sang Yong Song
- Department of Pathology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea
| | - Hee Jae Joo
- Department of Pathology, TCM Laboratory, Seongnam, Korea
| | - Sang-Yeop Yi
- Department of Pathology, Catholic Kwandong University College of Medicine, Gangneung, Korea
| | - Medical Informatics Study Group of the Korean Society of Pathologists
- Department of Hospital Pathology, College of Medicine, The Catholic University of Korea, Seoul, Korea
- Department of Pathology, Dong-A University College of Medicine, Busan, Korea
- Department of Pathology, Seoul Clinical Laboratories, Yongin, Korea
- Department of Pathology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea
- Department of Pathology, TCM Laboratory, Seongnam, Korea
- Department of Pathology, Catholic Kwandong University College of Medicine, Gangneung, Korea
| |
Collapse
|
4
|
Dlamini Z, Francies FZ, Hull R, Marima R. Artificial intelligence (AI) and big data in cancer and precision oncology. Comput Struct Biotechnol J 2020; 18:2300-2311. [PMID: 32994889 PMCID: PMC7490765 DOI: 10.1016/j.csbj.2020.08.019] [Citation(s) in RCA: 89] [Impact Index Per Article: 22.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/03/2020] [Revised: 08/21/2020] [Accepted: 08/21/2020] [Indexed: 02/07/2023] Open
Abstract
Artificial intelligence (AI) and machine learning have significantly influenced many facets of the healthcare sector. Advancement in technology has paved the way for analysis of big datasets in a cost- and time-effective manner. Clinical oncology and research are reaping the benefits of AI. The burden of cancer is a global phenomenon. Efforts to reduce mortality rates requires early diagnosis for effective therapeutic interventions. However, metastatic and recurrent cancers evolve and acquire drug resistance. It is imperative to detect novel biomarkers that induce drug resistance and identify therapeutic targets to enhance treatment regimes. The introduction of the next generation sequencing (NGS) platforms address these demands, has revolutionised the future of precision oncology. NGS offers several clinical applications that are important for risk predictor, early detection of disease, diagnosis by sequencing and medical imaging, accurate prognosis, biomarker identification and identification of therapeutic targets for novel drug discovery. NGS generates large datasets that demand specialised bioinformatics resources to analyse the data that is relevant and clinically significant. Through these applications of AI, cancer diagnostics and prognostic prediction are enhanced with NGS and medical imaging that delivers high resolution images. Regardless of the improvements in technology, AI has some challenges and limitations, and the clinical application of NGS remains to be validated. By continuing to enhance the progression of innovation and technology, the future of AI and precision oncology show great promise.
Collapse
Affiliation(s)
- Zodwa Dlamini
- SAMRC/UP Precision Prevention & Novel Drug Targets for HIV-Associated Cancers (PPNDTHAC) Extramural Unit, Pan African Cancer Research Institute (PACRI), University of Pretoria, Faculty of Health Sciences, Hatfield 0028, South Africa
| | - Flavia Zita Francies
- SAMRC/UP Precision Prevention & Novel Drug Targets for HIV-Associated Cancers (PPNDTHAC) Extramural Unit, Pan African Cancer Research Institute (PACRI), University of Pretoria, Faculty of Health Sciences, Hatfield 0028, South Africa
| | - Rodney Hull
- SAMRC/UP Precision Prevention & Novel Drug Targets for HIV-Associated Cancers (PPNDTHAC) Extramural Unit, Pan African Cancer Research Institute (PACRI), University of Pretoria, Faculty of Health Sciences, Hatfield 0028, South Africa
| | - Rahaba Marima
- SAMRC/UP Precision Prevention & Novel Drug Targets for HIV-Associated Cancers (PPNDTHAC) Extramural Unit, Pan African Cancer Research Institute (PACRI), University of Pretoria, Faculty of Health Sciences, Hatfield 0028, South Africa
| |
Collapse
|
5
|
Li HN, Wang RC, Chen JP, Chang ST, Chuang SS. Density and size of lymphoid follicles are useful clues in differentiating primary intestinal follicular lymphoma from intestinal reactive lymphoid hyperplasia. Diagn Pathol 2020; 15:82. [PMID: 32635930 PMCID: PMC7341590 DOI: 10.1186/s13000-020-00991-3] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/11/2020] [Accepted: 06/17/2020] [Indexed: 12/27/2022] Open
Abstract
Background Primary intestinal follicular lymphoma (PI-FL) is a rare and indolent lymphoma and is challenging for diagnosis with endoscopic biopsy specimens. Whole slide imaging (WSI) has been increasingly used for assisting pathologic diagnosis, but not for lymphoma yet, probably because there are usually too many immunostained sections in a single case. In this study we attempted to identify morphological clues of PI-FL in the endoscopic biopsy specimens by measuring various parameters using WSI. Methods We retrospectively investigated 21 PI-FL cases, and scanned the HE sections from 17 of these cases with endoscopic biopsy specimens. Sections from 17 intestinal biopsies showing reactive lymphoid hyperplasia were scanned for comparison. The density and diameter of lymphoid follicles and the shortest distance of these follicles to the surface epithelia were measured on WSI. Comparisons of the aforementioned parameters were made between the neoplastic and reactive follicles. Results The density of follicles was significantly higher in PI-FL than that of reactive hyperplasia (median 0.5 vs. 0.2/mm2; p < 0.01). Furthermore, the neoplastic follicles were significantly larger (median diameter 756.9 vs. 479.7 μm; p < 0.01). The shortest distance of follicles to the surface epithelia tended to be closer in PI-FL (104.7 vs. 177.8 μm, p = 0.056), but not statistically significant. Conclusions In this study we found that in PI-FL the density and diameter of lymphoid follicles as measured from WSI were significantly different from that of intestinal reactive lymphoid hyperplasia. When facing the diagnostic challenge between these two entities in routine practice, pathologists might be alerted by these morphological clues and request for immunohistochemistry for differential diagnosis.
Collapse
Affiliation(s)
- Hsin-Ni Li
- Department of Pathology and Laboratory Medicine, Taichung Veterans General Hospital, Taichung, Taiwan
| | - Ren Ching Wang
- Department of Pathology and Laboratory Medicine, Taichung Veterans General Hospital, Taichung, Taiwan.,Department of Nursing, HungKuang University, Taichung, Taiwan
| | - Jun-Peng Chen
- Biostatistics Task Force, Taichung Veterans General Hospital, Taichung, Taiwan
| | - Sheng-Tsung Chang
- Department of Pathology, Chi-Mei Medical Center, 901 Chung-Hwa Road, Yong-Kang District, Tainan, 71004, Taiwan.,Department of Nursing, National Tainan Institute of Nursing, Tainan, Taiwan
| | - Shih-Sung Chuang
- Department of Pathology, Chi-Mei Medical Center, 901 Chung-Hwa Road, Yong-Kang District, Tainan, 71004, Taiwan. .,Department of Pathology, School of Medicine, College of Medicine, National Taiwan University, Taipei, Taiwan.
| |
Collapse
|
6
|
Salama ME, Macon WR, Pantanowitz L. Is the Time Right to Start Using Digital Pathology and Artificial Intelligence for the Diagnosis of Lymphoma? J Pathol Inform 2020; 11:16. [PMID: 33033653 PMCID: PMC7513776 DOI: 10.4103/jpi.jpi_16_20] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/10/2020] [Revised: 04/01/2020] [Accepted: 04/13/2020] [Indexed: 12/17/2022] Open
Affiliation(s)
| | | | - Liron Pantanowitz
- University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, USA
| |
Collapse
|
7
|
Parwani AV. Next generation diagnostic pathology: use of digital pathology and artificial intelligence tools to augment a pathological diagnosis. Diagn Pathol 2019; 14:138. [PMID: 31881972 PMCID: PMC6933733 DOI: 10.1186/s13000-019-0921-2] [Citation(s) in RCA: 53] [Impact Index Per Article: 10.6] [Reference Citation Analysis] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/24/2022] Open
Affiliation(s)
- Anil V Parwani
- Division of Digital and Computational Pathology, Department of Pathology, The Ohio State University, Columbus, USA.
| |
Collapse
|