1
|
Coudry RA, Assis EA, Frassetto FP, Jansen AM, da Silva LM, Parra-Medina R, Saieg M. Crossing the Andes: Challenges and opportunities for digital pathology in Latin America. J Pathol Inform 2024; 15:100369. [PMID: 38638195 PMCID: PMC11025004 DOI: 10.1016/j.jpi.2024.100369] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/18/2023] [Revised: 02/05/2024] [Accepted: 02/17/2024] [Indexed: 04/20/2024] Open
Abstract
The most widely accepted and used type of digital pathology (DP) is whole-slide imaging (WSI). The USFDA granted two WSI system approvals for primary diagnosis, the first in 2017. In Latin America, DP has the potential to reshape healthcare by enhancing diagnostic capabilities through artificial intelligence (AI) and standardizing pathology reports. Yet, we must tackle regulatory hurdles, training, resource availability, and unique challenges to the region. Collectively addressing these hurdles can enable the region to harness DP's advantages-enhancing disease diagnosis, medical research, and healthcare accessibility for its population. Americas Health Foundation assembled a panel of Latin American pathologists who are experts in DP to assess the hurdles to implementing it into pathologists' workflows in the region and provide recommendations for overcoming them. Some key steps recommended include creating a Latin American Society of Digital Pathology to provide continuing education, developing AI models trained on the Latin American population, establishing national regulatory frameworks for protecting the data, and standardizing formats for DP images to ensure that pathologists can collaborate and validate specimens across the various DP platforms.
Collapse
Affiliation(s)
| | | | | | | | | | - Rafael Parra-Medina
- National Cancer Institute (INC), Bogotá, Colombia
- Fundación Universitaria de Ciencias de la Salud (FUCS), Bogotá, Colombia
| | - Mauro Saieg
- Grupo Fleury, São Paulo, Brazil
- Santa Casa Medical School, São Paulo, SP, Brazil
| |
Collapse
|
2
|
Moscalu M, Moscalu R, Dascălu CG, Țarcă V, Cojocaru E, Costin IM, Țarcă E, Șerban IL. Histopathological Images Analysis and Predictive Modeling Implemented in Digital Pathology-Current Affairs and Perspectives. Diagnostics (Basel) 2023; 13:2379. [PMID: 37510122 PMCID: PMC10378281 DOI: 10.3390/diagnostics13142379] [Citation(s) in RCA: 4] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/25/2023] [Revised: 07/11/2023] [Accepted: 07/12/2023] [Indexed: 07/30/2023] Open
Abstract
In modern clinical practice, digital pathology has an essential role, being a technological necessity for the activity in the pathological anatomy laboratories. The development of information technology has majorly facilitated the management of digital images and their sharing for clinical use; the methods to analyze digital histopathological images, based on artificial intelligence techniques and specific models, quantify the required information with significantly higher consistency and precision compared to that provided by optical microscopy. In parallel, the unprecedented advances in machine learning facilitate, through the synergy of artificial intelligence and digital pathology, the possibility of diagnosis based on image analysis, previously limited only to certain specialties. Therefore, the integration of digital images into the study of pathology, combined with advanced algorithms and computer-assisted diagnostic techniques, extends the boundaries of the pathologist's vision beyond the microscopic image and allows the specialist to use and integrate his knowledge and experience adequately. We conducted a search in PubMed on the topic of digital pathology and its applications, to quantify the current state of knowledge. We found that computer-aided image analysis has a superior potential to identify, extract and quantify features in more detail compared to the human pathologist's evaluating possibilities; it performs tasks that exceed its manual capacity, and can produce new diagnostic algorithms and prediction models applicable in translational research that are able to identify new characteristics of diseases based on changes at the cellular and molecular level.
Collapse
Affiliation(s)
- Mihaela Moscalu
- Department of Preventive Medicine and Interdisciplinarity, "Grigore T. Popa" University of Medicine and Pharmacy, 700115 Iassy, Romania
| | - Roxana Moscalu
- Wythenshawe Hospital, Manchester University NHS Foundation Trust, Manchester Academic Health Science Centre, Manchester M139PT, UK
| | - Cristina Gena Dascălu
- Department of Preventive Medicine and Interdisciplinarity, "Grigore T. Popa" University of Medicine and Pharmacy, 700115 Iassy, Romania
| | - Viorel Țarcă
- Department of Preventive Medicine and Interdisciplinarity, "Grigore T. Popa" University of Medicine and Pharmacy, 700115 Iassy, Romania
| | - Elena Cojocaru
- Department of Morphofunctional Sciences I, "Grigore T. Popa" University of Medicine and Pharmacy, 700115 Iassy, Romania
| | - Ioana Mădălina Costin
- Faculty of Medicine, "Grigore T. Popa" University of Medicine and Pharmacy, 700115 Iassy, Romania
| | - Elena Țarcă
- Department of Surgery II-Pediatric Surgery, "Grigore T. Popa" University of Medicine and Pharmacy, 700115 Iassy, Romania
| | - Ionela Lăcrămioara Șerban
- Department of Morpho-Functional Sciences II, Faculty of Medicine, "Grigore T. Popa" University of Medicine and Pharmacy, 700115 Iassy, Romania
| |
Collapse
|
3
|
Liscia DS, D’Andrea M, Biletta E, Bellis D, Demo K, Ferrero F, Petti A, Butinar R, D’Andrea E, Davini G. Use of digital pathology and artificial intelligence for the diagnosis of Helicobacter pylori in gastric biopsies. Pathologica 2022; 114:295-303. [DOI: 10.32074/1591-951x-751] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/22/2022] [Accepted: 01/22/2022] [Indexed: 12/20/2022] Open
|
4
|
Rizzo PC, Girolami I, Marletta S, Pantanowitz L, Antonini P, Brunelli M, Santonicco N, Vacca P, Tumino N, Moretta L, Parwani A, Satturwar S, Eccher A, Munari E. Technical and Diagnostic Issues in Whole Slide Imaging Published Validation Studies. Front Oncol 2022; 12:918580. [PMID: 35785212 PMCID: PMC9246412 DOI: 10.3389/fonc.2022.918580] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/12/2022] [Accepted: 05/24/2022] [Indexed: 01/07/2023] Open
Abstract
ObjectiveDigital pathology with whole-slide imaging (WSI) has many potential clinical and non-clinical applications. In the past two decades, despite significant advances in WSI technology adoption remains slow for primary diagnosis. The aim of this study was to identify common pitfalls of WSI reported in validation studies and offer measures to overcome these challenges.MethodsA systematic search was conducted in the electronic databases Pubmed-MEDLINE and Embase. Inclusion criteria were all validation studies designed to evaluate the feasibility of WSI for diagnostic clinical use in pathology. Technical and diagnostic problems encountered with WSI in these studies were recorded.ResultsA total of 45 studies were identified in which technical issues were reported in 15 (33%), diagnostic issues in 8 (18%), and 22 (49%) reported both. Key technical problems encompassed slide scan failure, prolonged time for pathologists to review cases, and a need for higher image resolution. Diagnostic challenges encountered were concerned with grading dysplasia, reliable assessment of mitoses, identification of microorganisms, and clearly defining the invasive front of tumors.ConclusionDespite technical advances with WSI technology, some critical concerns remain that need to be addressed to ensure trustworthy clinical diagnostic use. More focus on the quality of the pre-scanning phase and training of pathologists could help reduce the negative impact of WSI technical difficulties. WSI also seems to exacerbate specific diagnostic tasks that are already challenging among pathologists even when examining glass slides with conventional light microscopy.
Collapse
Affiliation(s)
- Paola Chiara Rizzo
- Department of Pathology and Diagnostics and Public Health, Section of Pathology, University Hospital of Verona, Verona, Italy
| | | | - Stefano Marletta
- Department of Pathology and Diagnostics and Public Health, Section of Pathology, University Hospital of Verona, Verona, Italy
- Department of Pathology, Pederzoli Hospital, Peschiera del Garda, Italy
| | - Liron Pantanowitz
- Department of Pathology & Clinical Labs, University of Michigan, Ann Arbor, MI, United States
| | - Pietro Antonini
- Department of Pathology and Diagnostics and Public Health, Section of Pathology, University Hospital of Verona, Verona, Italy
| | - Matteo Brunelli
- Department of Pathology and Diagnostics and Public Health, Section of Pathology, University Hospital of Verona, Verona, Italy
| | - Nicola Santonicco
- Department of Pathology and Diagnostics and Public Health, Section of Pathology, University Hospital of Verona, Verona, Italy
| | - Paola Vacca
- Bambino Gesù Children’s Hospital, Istituto di Ricovero e Cura a Carattere Scientifico (IRCCS), Rome, Italy
| | - Nicola Tumino
- Bambino Gesù Children’s Hospital, Istituto di Ricovero e Cura a Carattere Scientifico (IRCCS), Rome, Italy
| | - Lorenzo Moretta
- Bambino Gesù Children’s Hospital, Istituto di Ricovero e Cura a Carattere Scientifico (IRCCS), Rome, Italy
| | - Anil Parwani
- Department of Pathology, Ohio State University Medical Center, Columbus, OH, United States
| | - Swati Satturwar
- Department of Pathology, Ohio State University Medical Center, Columbus, OH, United States
| | - Albino Eccher
- Department of Pathology and Diagnostics, University and Hospital Trust of Verona, Verona, Italy
- *Correspondence: Albino Eccher,
| | - Enrico Munari
- Department of Molecular and Translational Medicine, University of Brescia, Brescia, Italy
| |
Collapse
|
5
|
Ferreira D, Vale J, Curado M, Polónia A, Eloy C. The impact of different coverslipping methods in the quality of the whole slide images used for diagnosis in pathology. J Pathol Inform 2022; 13:100098. [PMID: 36268095 PMCID: PMC9576983 DOI: 10.1016/j.jpi.2022.100098] [Citation(s) in RCA: 6] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 12/03/2021] [Indexed: 11/28/2022] Open
Abstract
Digital pathology workflow aims to create whole slide images (WSIs) for diagnosis. The quality of the WSIs depends primarily on the quality of the glass slides produced by the pathology laboratory, where the coverslipping method plays an important role. In this study we compare the glass, the film, and the liquid coverslipping methods to evaluate which ones are suitable to create WSIs for diagnosis. The study included 18 formalin-fixed paraffin-embedded tissue blocks. Of each block, 3 consecutive sections were covered using 1 of the 3 methods. The slides were scanned and evaluated for quality criteria by 2 pathologists experienced in digital pathology. The coverslipping method interferes with the quality of the WSIs, as well as with the scanning time and the file size of the WSIs. All coverslipping methods were found suitable for diagnosis. The glass and liquid methods were manual and had similar results concerning the presence of air bubbles/polymer accumulation, air drying artefacts, tissue exposed, and staining alterations. The glass method was the one with more air bubbles. The liquid method was associated with more alterations on the WSIs, but with the lowest file sizes. Automation of coverslipping and calibration of the scanner for the coverslipping method chosen by the pathology laboratory are relevant for the final quality of the WSIs.
Collapse
|
6
|
Evans AJ, Brown RW, Bui MM, Chlipala EA, Lacchetti C, Milner DA, Pantanowitz L, Parwani AV, Reid K, Riben MW, Reuter VE, Stephens L, Stewart RL, Thomas NE. Validating Whole Slide Imaging Systems for Diagnostic Purposes in Pathology. Arch Pathol Lab Med 2022; 146:440-450. [PMID: 34003251 DOI: 10.5858/arpa.2020-0723-cp] [Citation(s) in RCA: 72] [Impact Index Per Article: 36.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 03/26/2021] [Indexed: 11/06/2022]
Abstract
CONTEXT.— The original guideline, "Validating Whole Slide Imaging for Diagnostic Purposes in Pathology," was published in 2013 and included 12 guideline statements. The College of American Pathologists convened an expert panel to update the guideline following standards established by the National Academies of Medicine for developing trustworthy clinical practice guidelines. OBJECTIVE.— To assess evidence published since the release of the original guideline and provide updated recommendations for validating whole slide imaging (WSI) systems used for diagnostic purposes. DESIGN.— An expert panel performed a systematic review of the literature. Frozen sections, anatomic pathology specimens (biopsies, curettings, and resections), and hematopathology cases were included. Cytology cases were excluded. Using the Grading of Recommendations Assessment, Development, and Evaluation approach, the panel reassessed and updated the original guideline recommendations. RESULTS.— Three strong recommendations and 9 good practice statements are offered to assist laboratories with validating WSI digital pathology systems. CONCLUSIONS.— Systematic review of literature following release of the 2013 guideline reaffirms the use of a validation set of at least 60 cases, establishing intraobserver diagnostic concordance between WSI and glass slides and the use of a 2-week washout period between modalities. Although all discordances between WSI and glass slide diagnoses discovered during validation need to be reconciled, laboratories should be particularly concerned if their overall WSI-glass slide concordance is less than 95%.
Collapse
Affiliation(s)
- Andrew J Evans
- From the Department of Pathology, Mackenzie Health, Richmond Hill, Ontario, Canada (Evans)
| | - Richard W Brown
- The Department of Pathology, Memorial Hermann Southwest Hospital, Houston, Texas (Brown)
| | - Marilyn M Bui
- The Department of Pathology, Moffitt Cancer Center, Tampa, Florida (Bui)
| | | | - Christina Lacchetti
- Policy and Advocacy, American Society of Clinical Oncology, Alexandria, Virginia (Lacchetti)
| | - Danny A Milner
- American Society for Clinical Pathology, Chicago, Illinois (Milner)
| | - Liron Pantanowitz
- The Department of Pathology, University of Michigan, Ann Arbor (Pantanowitz)
| | - Anil V Parwani
- The Department of Pathology, Ohio State University Medical Center, Columbus (Parwani)
| | | | - Michael W Riben
- The Department of Pathology, University of Texas MD Anderson Cancer Center, Houston (Riben)
| | - Victor E Reuter
- The Department of Pathology, Memorial Sloan-Kettering Cancer Center, New York, New York (Reuter)
| | - Lisa Stephens
- The Department of Anatomic Pathology, Cleveland Clinic, Cleveland, Ohio (Stephens)
| | - Rachel L Stewart
- Janssen Research & Development, Spring House, Pennsylvania (Stewart)
| | - Nicole E Thomas
- Surveys (Thomas), College of American Pathologists, Northfield, Illinois
| |
Collapse
|
7
|
Mutter G, Milstone D, Hwang D, Siegmund S, Bruce A. Measuring digital pathology throughput and tissue dropouts. J Pathol Inform 2022; 13:8. [PMID: 35136675 PMCID: PMC8794031 DOI: 10.4103/jpi.jpi_5_21] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/19/2021] [Revised: 06/05/2021] [Accepted: 06/20/2021] [Indexed: 11/04/2022] Open
Abstract
Background: Digital pathology operations that precede viewing by a pathologist have a substantial impact on costs and fidelity of the digital image. Scan time and file size determine throughput and storage costs, whereas tissue omission during digital capture (“dropouts”) compromises downstream interpretation. We compared how these variables differ across scanners. Methods: A 212 slide set randomly selected from a gynecologic-gestational pathology practice was used to benchmark scan time, file size, and image completeness. Workflows included the Hamamatsu S210 scanner (operated under default and optimized profiles) and the Leica GT450. Digital tissue dropouts were detected by the aligned overlay of macroscopic glass slide camera images (reference) with images created by the slide scanners whole slide images. Results: File size and scan time were highly correlated within each platform. Differences in GT450, default S210, and optimized S210 performance were seen in average file size (1.4 vs. 2.5 vs. 3.4 GB) and scan time (93 vs. 376 vs. 721 s). Dropouts were seen in 29.5% (186/631) of successful scans overall: from a low of 13.7% (29/212) for the optimized S210 profile, followed by 34.6% (73/211) for the GT450 and 40.4% (84/208) for the default profile S210 profile. Small dislodged fragments, “shards,” were dropped in 22.2% (140/631) of slides, followed by tissue marginalized at the glass slide edges, 6.2% (39/631). “Unique dropouts,” those for which no equivalent appeared elsewhere in the scan, occurred in only three slides. Of these, 67% (2/3) were “floaters” or contaminants from other cases. Conclusions: Scanning speed and resultant file size vary greatly by scanner type, scanner operation settings, and clinical specimen mix (tissue type, tissue area). Digital image fidelity as measured by tissue dropout frequency and dropout type also varies according to the tissue type and scanner. Dropped tissues very rarely (1/631) represent actual specimen tissues that are not represented elsewhere in the scan, so in most cases cannot alter the diagnosis. Digital pathology platforms vary in their output efficiency and image fidelity to the glass original and should be matched to the intended application.
Collapse
|
8
|
Jiang H, Yang Y, Qian Y, Shao C, Lu J, Bian Y, Zheng J. Tumor Budding Score Is a Strong and Independent Prognostic Factor in Patients With Pancreatic Ductal Adenocarcinoma: An Evaluation of Whole Slide Pathology Images of Large Sections. Front Oncol 2021; 11:740212. [PMID: 34917500 PMCID: PMC8668607 DOI: 10.3389/fonc.2021.740212] [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: 07/12/2021] [Accepted: 11/08/2021] [Indexed: 12/09/2022] Open
Abstract
OBJECTIVE We aimed to develop the tumor budding (TB) score and to explore the association between the TB score and overall survival (OS) in patients with pancreatic ductal adenocarcinoma (PDAC). METHODS In this retrospective study, 130 consecutive patients with PDAC underwent surgical resection between July 2016 and March 2019. The location and counts of TB were assessed based on the digitalized whole slide hematoxylin and eosin images. The TB score was achieved using the Cox regression equation. The cutoff point for the TB score was determined by X-tile. Univariate and multivariate Cox regression models were used to analyze the association between the TB score and OS. RESULTS The TB score was 0.49 (range = 0-1.08), and the best cutoff for the TB score was 0.62. The duration of survival in individuals with a low TB score [median = 21.8 months, 95% confidence interval (CI) = 15.43-25.50] was significantly longer than that in those with a high TB score (median = 11.33 months, 95% CI = 9.8-14.22). Univariate analysis revealed that the TB score was significantly associated with OS [hazard ratio (HR) = 2.71, 95% CI = 1.48-4.96, p = 0.001]. Multivariate analysis revealed a strong and independent association between the TB score and OS (HR = 2.35, 95% CI = 1.27-4.33, p = 0.03). The high TB score group had a 2.14 times higher mortality than the low TB score group. CONCLUSION The TB score is strongly and independently associated with the risk of OS in PDAC.
Collapse
Affiliation(s)
- Hui Jiang
- Department of Pathology, Changhai Hospital, The Naval Military Medical University, Shanghai, China
| | - Yelin Yang
- Department of Pathology, Changhai Hospital, The Naval Military Medical University, Shanghai, China
| | - Yuping Qian
- Department of Pathology, Changhai Hospital, The Naval Military Medical University, Shanghai, China
| | - Chengwei Shao
- Department of Radiology, Changhai Hospital, Naval Medical University, Shanghai, China
| | - Jianping Lu
- Department of Radiology, Changhai Hospital, Naval Medical University, Shanghai, China
| | - Yun Bian
- Department of Radiology, Changhai Hospital, Naval Medical University, Shanghai, China
| | - Jianming Zheng
- Department of Pathology, Changhai Hospital, The Naval Military Medical University, Shanghai, China
| |
Collapse
|
9
|
Shi W, Georgiou P, Akram A, Proute MC, Serhiyenia T, Kerolos ME, Pradeep R, Kothur NR, Khan S. Diagnostic Pitfalls of Digital Microscopy Versus Light Microscopy in Gastrointestinal Pathology: A Systematic Review. Cureus 2021; 13:e17116. [PMID: 34548958 PMCID: PMC8437006 DOI: 10.7759/cureus.17116] [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: 07/22/2021] [Accepted: 08/11/2021] [Indexed: 12/15/2022] Open
Abstract
Digital microscopy (DM) is one of the cutting-edge advances in pathology, which entails improved efficiency, diagnostic advantages, and potential application in virtual diagnosis, particularly in the current era of the coronavirus disease (COVID-19) pandemic. However, the diagnostic challenges are the remaining concerns for its wider adoption by pathologists, and these concerns should be addressed in a specific subspecialty. We aim to identify the common diagnostic pitfalls of whole slide imaging (WSI), one modality of DM, in gastrointestinal (GI) pathology. From validating studies of primary diagnosis performance, we included 16 records with features on GI cases involved, at least two weeks wash-out periods, and more than 60 case study designs. A tailored quality appraisal assessment was utilized to evaluate the risks of bias for these diagnostic accuracy studies. Furthermore, due to the highly heterogeneous studies and unstandardized definition of discordance, we extract the discordant cases in GI pathology and calculate the discrepant rate, resulting from 0.5% to 64.28%. Targeting discrepancy cases between digital microscopy and light microscopy, we demonstrate five main diagnostic pitfalls regarding WSI as follows: additional time to review slides in WSI, hard to identify dysplasia nucleus, missed organisms like Helicobacter pylori (H. pylori), specific cell recognitions, and technical issues. After detailed reviews and analysis, we generate two essential suggestions for further GI cases signing out by DM. One is to use systematized 20x scans for diagnostic workouts and requesting 40x or even 60x scans for challenging cases; another is that a high-volume slides training should be set before the real clinical application of WSI for primary diagnosis, particularly in GI pathology.
Collapse
Affiliation(s)
- Wangpan Shi
- Pathology, California Institute of Behavioral Neurosciences & Psychology, Fairfield, USA
| | - Petros Georgiou
- Internal Medicine, California Institute of Behavioral Neurosciences & Psychology, Fairfield, USA
- Department of Oncology, Oxford University, Oxford, GBR
| | - Aqsa Akram
- Internal Medicine, California Institute of Behavioral Neurosciences & Psychology, Fairfield, USA
| | - Matthew C Proute
- Internal Medicine, California Institute of Behavioral Neurosciences & Psychology, Fairfield, USA
| | - Tatsiana Serhiyenia
- Internal Medicine, California Institute of Behavioral Neurosciences & Psychology, Fairfield, USA
| | - Mina E Kerolos
- General Medicine, California Institute of Behavioral Neurosciences & Psychology, Fairfield, USA
| | - Roshini Pradeep
- Internal Medicine, California Institute of Behavioral Neurosciences & Psychology, Fairfield, USA
| | - Nageshwar R Kothur
- Internal Medicine, California Institute of Behavioral Neurosciences & Psychology, Fairfield, USA
| | - Safeera Khan
- Internal Medicine, California Institute of Behavioral Neurosciences & Psychology, Fairfield, USA
| |
Collapse
|
10
|
Azam AS, Miligy IM, Kimani PKU, Maqbool H, Hewitt K, Rajpoot NM, Snead DRJ. Diagnostic concordance and discordance in digital pathology: a systematic review and meta-analysis. J Clin Pathol 2021; 74:448-455. [PMID: 32934103 PMCID: PMC8223673 DOI: 10.1136/jclinpath-2020-206764] [Citation(s) in RCA: 35] [Impact Index Per Article: 11.7] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/09/2020] [Accepted: 07/10/2020] [Indexed: 01/10/2023]
Abstract
BACKGROUND Digital pathology (DP) has the potential to fundamentally change the way that histopathology is practised, by streamlining the workflow, increasing efficiency, improving diagnostic accuracy and facilitating the platform for implementation of artificial intelligence-based computer-assisted diagnostics. Although the barriers to wider adoption of DP have been multifactorial, limited evidence of reliability has been a significant contributor. A meta-analysis to demonstrate the combined accuracy and reliability of DP is still lacking in the literature. OBJECTIVES We aimed to review the published literature on the diagnostic use of DP and to synthesise a statistically pooled evidence on safety and reliability of DP for routine diagnosis (primary and secondary) in the context of validation process. METHODS A comprehensive literature search was conducted through PubMed, Medline, EMBASE, Cochrane Library and Google Scholar for studies published between 2013 and August 2019. The search protocol identified all studies comparing DP with light microscopy (LM) reporting for diagnostic purposes, predominantly including H&E-stained slides. Random-effects meta-analysis was used to pool evidence from the studies. RESULTS Twenty-five studies were deemed eligible to be included in the review which examined a total of 10 410 histology samples (average sample size 176). For overall concordance (clinical concordance), the agreement percentage was 98.3% (95% CI 97.4 to 98.9) across 24 studies. A total of 546 major discordances were reported across 25 studies. Over half (57%) of these were related to assessment of nuclear atypia, grading of dysplasia and malignancy. These were followed by challenging diagnoses (26%) and identification of small objects (16%). CONCLUSION The results of this meta-analysis indicate equivalent performance of DP in comparison with LM for routine diagnosis. Furthermore, the results provide valuable information concerning the areas of diagnostic discrepancy which may warrant particular attention in the transition to DP.
Collapse
Affiliation(s)
- Ayesha S Azam
- Cellular Pathology, University Hospitals Coventry and Warwickshire NHS Trust, Coventry, Coventry, UK
- Tissue Image Analytics Laboratory, Department of Computer Science, University of Warwick, Coventry, West Midlands, UK
| | - Islam M Miligy
- Nottingham Breast Cancer Research Centre (NBCRC), School of Medicine, University of Nottingham, Nottingham, Nottinghamshire, UK
| | - Peter K-U Kimani
- Warwick Medical School, University of Warwick, Coventry, West Midlands, UK
| | - Heeba Maqbool
- Cellular Pathology, University Hospitals Coventry and Warwickshire NHS Trust, Coventry, Coventry, UK
| | - Katherine Hewitt
- Cellular Pathology, University Hospitals Coventry and Warwickshire NHS Trust, Coventry, Coventry, UK
| | - Nasir M Rajpoot
- Tissue Image Analytics Laboratory, Department of Computer Science, University of Warwick, Coventry, West Midlands, UK
| | - David R J Snead
- Cellular Pathology, University Hospitals Coventry and Warwickshire NHS Trust, Coventry, Coventry, UK
| |
Collapse
|
11
|
Gavrielides MA, Ronnett BM, Vang R, Sheikhzadeh F, Seidman JD. Selection of Representative Histologic Slides in Interobserver Reproducibility Studies: Insights from Expert Review for Ovarian Carcinoma Subtype Classification. J Pathol Inform 2021; 12:15. [PMID: 34012719 PMCID: PMC8112350 DOI: 10.4103/jpi.jpi_56_20] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/25/2020] [Revised: 09/02/2020] [Accepted: 10/28/2020] [Indexed: 11/04/2022] Open
Abstract
BACKGROUND Observer studies in pathology often utilize a limited number of representative slides per case, selected and reported in a nonstandardized manner. Reference diagnoses are commonly assumed to be generalizable to all slides of a case. We examined these issues in the context of pathologist concordance for histologic subtype classification of ovarian carcinomas (OCs). MATERIALS AND METHODS A cohort of 114 OCs consisting of 72 cases with a single representative slide (Group 1) and 42 cases with multiple representative slides (148 slides, 2-6 sections per case, Group 2) was independently reviewed by three experts in gynecologic pathology (case-based review). In a follow-up study, each individual slide was independently reviewed in a randomized order by the same pathologists (section-based review). RESULTS Average interobserver concordance varied from 100% for Group 1 to 64.3% for Group 2 (86.8% across all cases). Across Group 2, 19 cases (45.2%) had at least one slide classified as a different subtype than the subtype assigned from case-based review, demonstrating the impact of intratumoral heterogeneity. Section-based concordance across individual sections from Group 2 was comparable to case-based concordance for those cases indicating diagnostic challenges at the individual section level. Findings demonstrate the increased diagnostic complexity of heterogeneous tumors that require multiple section sampling and its impact on pathologist performance. CONCLUSIONS The proportion of cases with multiple representative slides in cohorts used in validation studies, such as those conducted to evaluate artificial intelligence/machine learning tools, can influence diagnostic performance, and if not accounted for, can cause disparities between research and real-world observations and between research studies. Case selection in validation studies should account for tumor heterogeneity to create balanced datasets in terms of diagnostic complexity.
Collapse
Affiliation(s)
- Marios A. Gavrielides
- Division of Imaging, Diagnostics, and Software Reliability, Office of Science and Engineering Laboratories, Center for Devices and Radiological Health, U.S. Food and Drug Administration, Silver Spring, Maryland, USA, (Currently at AstraZeneca, Precision Medicine and Biosamples, Gaithersburg, Maryland, USA)
| | - Brigitte M. Ronnett
- Department of Pathology and Gynecology and Obstetrics, The Johns Hopkins Hospital, Baltimore, Maryland, USA
| | - Russell Vang
- Department of Pathology and Gynecology and Obstetrics, The Johns Hopkins Hospital, Baltimore, Maryland, USA
| | - Fahime Sheikhzadeh
- Electrical and Computer Engineering Department, University of British Columbia, Vancouver, Canada, (Currently at Roche Diagnostics, San Francisco, California, USA)
| | - Jeffrey D Seidman
- Division of Molecular Genetics and Pathology, Office of In Vitro Diagnostics and Radiological Health, Office of Product Evaluation and Quality, Center for Devices and Radiological Health, U.S. Food and Drug Administration, Silver Spring, Maryland, USA
| |
Collapse
|
12
|
Redemann J, Schultz FA, Martinez C, Harrell M, Clark DP, Martin DR, Hanson JA. Comparing Deep Learning and Immunohistochemistry in Determining the Site of Origin for Well-Differentiated Neuroendocrine Tumors. J Pathol Inform 2020; 11:32. [PMID: 33343993 PMCID: PMC7737494 DOI: 10.4103/jpi.jpi_37_20] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/23/2020] [Revised: 07/25/2020] [Accepted: 08/25/2020] [Indexed: 12/18/2022] Open
Abstract
Background Determining the site of origin for metastatic well-differentiated neuroendocrine tumors (WDNETs) is challenging, and immunohistochemical (IHC) profiles do not always lead to a definitive diagnosis. We sought to determine if a deep-learning convolutional neural network (CNN) could improve upon established IHC profiles in predicting the site of origin in a cohort of WDNETs from the common primary sites. Materials and Methods Hematoxylin and eosin (H&E)-stained tissue microarrays (TMAs) were created using 215 WDNETs arising from the known primary sites. A CNN trained and tested on 60% (n = 130) and 40% (n = 85) of these cases, respectively. One hundred and seventy-nine cases had TMA tissue remaining for the IHC analysis. These cases were stained with IHC markers pPAX8, CDX2, SATB2, and thyroid transcription factor-1 (markers of pancreas/duodenum, ileum/jejunum/duodenum, colorectum/appendix, and lung WDNET sites of origin, respectively). The CNN diagnosis was deemed correct if it designated a majority or plurality of the tumor area as the known site of origin. The IHC diagnosis was deemed correct if the most specific marker for a particular site of origin met an H-score threshold determined by two pathologists. Results When all cases were considered, the CNN correctly identified the site of origin at a lower rate compared to IHC (72% vs. 82%, respectively). Of the 85 cases in the CNN test set, 66 had sufficient TMA material for IHC stains, thus 66 cases were available for a direct case-by-case comparison of IHC versus CNN. The CNN correctly identified 70% of these cases, while IHC correctly identified 76%, a finding that was not statistically significant (P = 0.56). Conclusion A CNN can identify WDNET site of origin at an accuracy rate close to the current gold standard IHC methods.
Collapse
Affiliation(s)
- Jordan Redemann
- Department of Pathology, University of New Mexico School of Medicine, Albuquerque, NM, USA
| | - Fred A Schultz
- Department of Pathology, University of New Mexico School of Medicine, Albuquerque, NM, USA
| | - Cathy Martinez
- Department of Pathology, University of New Mexico School of Medicine, Albuquerque, NM, USA
| | - Michael Harrell
- Department of Pathology, University of New Mexico School of Medicine, Albuquerque, NM, USA
| | - Douglas P Clark
- Department of Pathology, University of New Mexico School of Medicine, Albuquerque, NM, USA
| | - David R Martin
- Department of Pathology, University of New Mexico School of Medicine, Albuquerque, NM, USA
| | - Joshua A Hanson
- Department of Pathology, University of New Mexico School of Medicine, Albuquerque, NM, USA
| |
Collapse
|
13
|
Niazi MKK, Parwani AV, Gurcan MN. Digital pathology and artificial intelligence. Lancet Oncol 2019; 20:e253-e261. [PMID: 31044723 PMCID: PMC8711251 DOI: 10.1016/s1470-2045(19)30154-8] [Citation(s) in RCA: 505] [Impact Index Per Article: 101.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/29/2019] [Revised: 02/28/2019] [Accepted: 03/13/2019] [Indexed: 02/06/2023]
Abstract
In modern clinical practice, digital pathology has a crucial role and is increasingly a technological requirement in the scientific laboratory environment. The advent of whole-slide imaging, availability of faster networks, and cheaper storage solutions has made it easier for pathologists to manage digital slide images and share them for clinical use. In parallel, unprecedented advances in machine learning have enabled the synergy of artificial intelligence and digital pathology, which offers image-based diagnosis possibilities that were once limited only to radiology and cardiology. Integration of digital slides into the pathology workflow, advanced algorithms, and computer-aided diagnostic techniques extend the frontiers of the pathologist's view beyond a microscopic slide and enable true utilisation and integration of knowledge that is beyond human limits and boundaries, and we believe there is clear potential for artificial intelligence breakthroughs in the pathology setting. In this Review, we discuss advancements in digital slide-based image diagnosis for cancer along with some challenges and opportunities for artificial intelligence in digital pathology.
Collapse
Affiliation(s)
| | - Anil V Parwani
- Department of Pathology, The Ohio State University, Columbus, OH, USA
| | - Metin N Gurcan
- Center for Biomedical Informatics, Wake Forest School of Medicine, Winston-Salem, NC, USA
| |
Collapse
|
14
|
The performance of digital microscopy for primary diagnosis in human pathology: a systematic review. Virchows Arch 2019; 474:269-287. [DOI: 10.1007/s00428-018-02519-z] [Citation(s) in RCA: 30] [Impact Index Per Article: 6.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/09/2018] [Revised: 12/25/2018] [Accepted: 12/28/2018] [Indexed: 02/06/2023]
|
15
|
Validation of digital microscopy in the histopathological diagnoses of oral diseases. Virchows Arch 2018; 473:321-327. [DOI: 10.1007/s00428-018-2382-5] [Citation(s) in RCA: 9] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/17/2018] [Accepted: 05/21/2018] [Indexed: 01/17/2023]
|
16
|
Villa I, Mathieu MC, Bosq J, Auperin A, Pomerol JF, Lacroix-Triki M, Scoazec JY, Dartigues P. Daily Biopsy Diagnosis in Surgical Pathology: Concordance Between Light Microscopy and Whole-Slide Imaging in Real-Life Conditions. Am J Clin Pathol 2018; 149:344-351. [PMID: 29452345 DOI: 10.1093/ajcp/aqx161] [Citation(s) in RCA: 8] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/04/2023] Open
Abstract
OBJECTIVES The current challenge for the various digital whole-slide imaging (WSI) systems is to be definitively validated for diagnostic purposes. We designed a concordance study between glass slide and digital slide diagnosis in real-life conditions, coupled with an ergonomic study. METHODS Three senior pathologists evaluated, first in glass slides and then in digital slides, 119 biopsy cases, including 749 slides, with 332 H&E saffron stains and 417 additional techniques, mainly immunohistochemistry. RESULTS All digital slides, including specially stained slides, were interpretable. Concordance between glass slides and digital slides was observed in 87.4% of cases. Minor discordances were observed in 12 (10.1%) cases and major discordances, with therapeutic impact, in three (2.5%), including one related to WSI. The satisfaction of participants was high and increased with time. CONCLUSIONS Our study confirms the feasibility and accuracy of WSI diagnosis, even for cases having multiple samples and requiring special staining techniques, such as immunohistochemistry and in situ hybridization.
Collapse
Affiliation(s)
- Irène Villa
- Département de Biologie et Pathologie Médicales, Service de Pathologie Morphologique, Villejuif, France
| | - Marie-Christine Mathieu
- Département de Biologie et Pathologie Médicales, Service de Pathologie Morphologique, Villejuif, France
| | - Jacques Bosq
- Département de Biologie et Pathologie Médicales, Service de Pathologie Morphologique, Villejuif, France
| | - Anne Auperin
- Service de Biostatistique et d’Epidémiologie, Gustave Roussy Cancer Campus, Villejuif, France
| | | | - Magali Lacroix-Triki
- Département de Biologie et Pathologie Médicales, Service de Pathologie Morphologique, Villejuif, France
| | - Jean-Yves Scoazec
- Département de Biologie et Pathologie Médicales, Service de Pathologie Morphologique, Villejuif, France
- Faculté de Médecine de Bicêtre, Université Paris Saclay, Université Paris Sud XI, Le Kremlin-Bicêtre, France
| | - Peggy Dartigues
- Département de Biologie et Pathologie Médicales, Service de Pathologie Morphologique, Villejuif, France
| |
Collapse
|
17
|
Whole Slide Imaging Versus Microscopy for Primary Diagnosis in Surgical Pathology: A Multicenter Blinded Randomized Noninferiority Study of 1992 Cases (Pivotal Study). Am J Surg Pathol 2017; 42:39-52. [PMID: 28961557 PMCID: PMC5737464 DOI: 10.1097/pas.0000000000000948] [Citation(s) in RCA: 228] [Impact Index Per Article: 32.6] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/06/2023]
Abstract
Most prior studies of primary diagnosis in surgical pathology using whole slide imaging (WSI) versus microscopy have focused on specific organ systems or included relatively few cases. The objective of this study was to demonstrate that WSI is noninferior to microscopy for primary diagnosis in surgical pathology. A blinded randomized noninferiority study was conducted across the entire range of surgical pathology cases (biopsies and resections, including hematoxylin and eosin, immunohistochemistry, and special stains) from 4 institutions using the original sign-out diagnosis (baseline diagnosis) as the reference standard. Cases were scanned, converted to WSI and randomized. Sixteen pathologists interpreted cases by microscopy or WSI, followed by a wash-out period of ≥4 weeks, after which cases were read by the same observers using the other modality. Major discordances were identified by an adjudication panel, and the differences between major discordance rates for both microscopy (against the reference standard) and WSI (against the reference standard) were calculated. A total of 1992 cases were included, resulting in 15,925 reads. The major discordance rate with the reference standard diagnosis was 4.9% for WSI and 4.6% for microscopy. The difference between major discordance rates for microscopy and WSI was 0.4% (95% confidence interval, -0.30% to 1.01%). The difference in major discordance rates for WSI and microscopy was highest in endocrine pathology (1.8%), neoplastic kidney pathology (1.5%), urinary bladder pathology (1.3%), and gynecologic pathology (1.2%). Detailed analysis of these cases revealed no instances where interpretation by WSI was consistently inaccurate compared with microscopy for multiple observers. We conclude that WSI is noninferior to microscopy for primary diagnosis in surgical pathology, including biopsies and resections stained with hematoxylin and eosin, immunohistochemistry and special stains. This conclusion is valid across a wide variety of organ systems and specimen types.
Collapse
|
18
|
Evans AJ, Salama ME, Henricks WH, Pantanowitz L. Implementation of Whole Slide Imaging for Clinical Purposes: Issues to Consider From the Perspective of Early Adopters. Arch Pathol Lab Med 2017; 141:944-959. [PMID: 28440660 DOI: 10.5858/arpa.2016-0074-oa] [Citation(s) in RCA: 72] [Impact Index Per Article: 10.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/06/2022]
Abstract
CONTEXT - There is growing interest in the use of digital pathology, especially whole slide imaging, for diagnostic purposes. Many issues need to be considered when incorporating this technology into a clinical laboratory. The College of American Pathologists (CAP) established a Digital Pathology Committee to support the development of CAP programs related to digital pathology. One of its many initiatives was a panel discussion entitled "Implementing Whole-Slide Imaging for Clinical Use: What to Do and What to Avoid," given for 3 years at the CAP annual meetings starting in 2014. OBJECTIVES - To review major issues to consider when implementing whole slide imaging for clinical purposes as covered during the panel discussion. DESIGN - The views expressed and recommendations given are based primarily on the personal experience of the authors as early adopters of this technology. It is not intended to be an exhaustive review of digital pathology. RESULTS - Implementation is best approached in phases. Early efforts are directed toward identifying initial clinical applications and assembling an implementation team. Scanner selection should be based on intended use and budget. Recognizing pathologist concerns over the use of digital pathology for diagnostic purposes, ensuring adequate training, and performing appropriate validation studies will enhance adoption. Once implemented, the transition period from glass slide to image-based diagnostics will be associated with challenges, especially those related to a hybrid glass slide-digital slide workflow. CONCLUSIONS - With appropriate preparation, planning, and stepwise implementation, whole slide imaging can be used safely and reliably for frozen sections, consultation, quality assurance, and primary diagnosis.
Collapse
Affiliation(s)
| | | | | | - Liron Pantanowitz
- From the Department of Pathology, University Health Network, Toronto, Ontario, Canada (Dr Evans); the Department of Pathology, University of Utah and ARUP Laboratories, Reference Laboratory, Salt Lake City (Dr Salama); the Department of Pathology, Cleveland Clinic Foundation, Cleveland, Ohio (Dr Henricks); and the Department of Pathology, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania (Dr Pantanowitz)
| |
Collapse
|
19
|
Williams BJ, DaCosta P, Goacher E, Treanor D. A Systematic Analysis of Discordant Diagnoses in Digital Pathology Compared With Light Microscopy. Arch Pathol Lab Med 2017; 141:1712-1718. [PMID: 28467215 DOI: 10.5858/arpa.2016-0494-oa] [Citation(s) in RCA: 48] [Impact Index Per Article: 6.9] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/06/2022]
Abstract
CONTEXT - Relatively little is known about the significance and potential impact of glass-digital discordances, and this is likely to be of importance when considering digital pathology adoption. OBJECTIVE - To apply evidence-based medicine to collect and analyze reported instances of glass-digital discordance from the whole slide imaging validation literature. DESIGN - We used our prior systematic review protocol to identify studies assessing the concordance of light microscopy and whole slide imaging between 1999 and 2015. Data were extracted and analyzed by a team of histopathologists to classify the type, significance, and potential root cause of discordances. RESULTS - Twenty-three studies were included, yielding 8069 instances of a glass diagnosis being compared with a digital diagnosis. From these 8069 comparisons, 335 instances of discordance (4%) were reported, in which glass was the preferred diagnostic medium in 286 (85%), and digital in 44 (13%), with no consensus in 5 (2%). Twenty-eight discordances had the potential to cause moderate/severe patient harm. Of these, glass was the preferred diagnostic medium for 26 (93%). Of the 335 discordances, 109 (32%) involved the diagnosis or grading of dysplasia. For these cases, glass was the preferred diagnostic medium in 101 cases (93%), suggesting that diagnosis and grading of dysplasia may be a potential pitfall of digital diagnosis. In 32 of 335 cases (10%), discordance on digital was attributed to the inability to find a small diagnostic/prognostic object. CONCLUSIONS - Systematic analysis of concordance studies reveals specific areas that may be problematic on whole slide imaging. It is important that pathologists are aware of these areas to ensure patient safety.
Collapse
|
20
|
Chaddad A, Desrosiers C, Bouridane A, Toews M, Hassan L, Tanougast C. Multi Texture Analysis of Colorectal Cancer Continuum Using Multispectral Imagery. PLoS One 2016; 11:e0149893. [PMID: 26901134 PMCID: PMC4764026 DOI: 10.1371/journal.pone.0149893] [Citation(s) in RCA: 37] [Impact Index Per Article: 4.6] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/27/2015] [Accepted: 02/05/2016] [Indexed: 01/05/2023] Open
Abstract
PURPOSE This paper proposes to characterize the continuum of colorectal cancer (CRC) using multiple texture features extracted from multispectral optical microscopy images. Three types of pathological tissues (PT) are considered: benign hyperplasia, intraepithelial neoplasia and carcinoma. MATERIALS AND METHODS In the proposed approach, the region of interest containing PT is first extracted from multispectral images using active contour segmentation. This region is then encoded using texture features based on the Laplacian-of-Gaussian (LoG) filter, discrete wavelets (DW) and gray level co-occurrence matrices (GLCM). To assess the significance of textural differences between PT types, a statistical analysis based on the Kruskal-Wallis test is performed. The usefulness of texture features is then evaluated quantitatively in terms of their ability to predict PT types using various classifier models. RESULTS Preliminary results show significant texture differences between PT types, for all texture features (p-value < 0.01). Individually, GLCM texture features outperform LoG and DW features in terms of PT type prediction. However, a higher performance can be achieved by combining all texture features, resulting in a mean classification accuracy of 98.92%, sensitivity of 98.12%, and specificity of 99.67%. CONCLUSIONS These results demonstrate the efficiency and effectiveness of combining multiple texture features for characterizing the continuum of CRC and discriminating between pathological tissues in multispectral images.
Collapse
Affiliation(s)
- Ahmad Chaddad
- Laboratory for Imagery, Vision and Artificial Intelligence, École de Technologie Supérieure, Montréal, Québec, Canada
- Laboratory of Conception, Optimization and Modelling of Systems, University of Lorraine, Metz, Lorraine, France
| | - Christian Desrosiers
- Laboratory for Imagery, Vision and Artificial Intelligence, École de Technologie Supérieure, Montréal, Québec, Canada
| | - Ahmed Bouridane
- School of Computing, Engineering and Information Sciences, Northumbria University, Newcastle, United Kingdom
| | - Matthew Toews
- Laboratory for Imagery, Vision and Artificial Intelligence, École de Technologie Supérieure, Montréal, Québec, Canada
| | - Lama Hassan
- Laboratory of Conception, Optimization and Modelling of Systems, University of Lorraine, Metz, Lorraine, France
| | - Camel Tanougast
- Laboratory of Conception, Optimization and Modelling of Systems, University of Lorraine, Metz, Lorraine, France
| |
Collapse
|