Publications by authors named "Liron Pantanowitz"

452 Publications

The Ethics of Artificial Intelligence in Pathology and Laboratory Medicine: Principles and Practice.

Acad Pathol 2021 Jan-Dec;8:2374289521990784. Epub 2021 Feb 16.

Department of Pathology, University of Michigan, Ann Arbor, MI, USA.

Growing numbers of artificial intelligence applications are being developed and applied to pathology and laboratory medicine. These technologies introduce risks and benefits that must be assessed and managed through the lens of ethics. This article describes how long-standing principles of medical and scientific ethics can be applied to artificial intelligence using examples from pathology and laboratory medicine.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1177/2374289521990784DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7894680PMC
February 2021

Machine learning and augmented human intelligence use in histomorphology for haematolymphoid disorders.

Pathology 2021 Feb 25. Epub 2021 Feb 25.

Division of Hematopathology, Mayo Clinic, Rochester, MN, USA. Electronic address:

Advances in digital pathology have allowed a number of opportunities such as decision support using artificial intelligence (AI). The application of AI to digital pathology data shows promise as an aid for pathologists in the diagnosis of haematological disorders. AI-based applications have embraced benign haematology, diagnosing leukaemia and lymphoma, as well as ancillary testing modalities including flow cytometry. In this review, we highlight the progress made to date in machine learning applications in haematopathology, summarise important studies in this field, and highlight key limitations. We further present our outlook on the future direction and trends for AI to support diagnostic decisions in haematopathology.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.pathol.2020.12.004DOI Listing
February 2021

AI reality check when evaluating difficult to grade prostate cancers.

Virchows Arch 2021 Feb 2. Epub 2021 Feb 2.

Department of Pathology, University of Michigan, NCRC Bldg 35, 2800 Plymouth Road, Ann Arbor, MI, 48109-2800, USA.

View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1007/s00428-021-03045-1DOI Listing
February 2021

Artificial Intelligence-Based Screening for Mycobacteria in Whole-Slide Images of Tissue Samples.

Am J Clin Pathol 2021 Feb 2. Epub 2021 Feb 2.

aetherAI, Taipei, Taiwan.

Objectives: This study aimed to develop and validate a deep learning algorithm to screen digitized acid fast-stained (AFS) slides for mycobacteria within tissue sections.

Methods: A total of 441 whole-slide images (WSIs) of AFS tissue material were used to develop a deep learning algorithm. Regions of interest with possible acid-fast bacilli (AFBs) were displayed in a web-based gallery format alongside corresponding WSIs for pathologist review. Artificial intelligence (AI)-assisted analysis of another 138 AFS slides was compared to manual light microscopy and WSI evaluation without AI support.

Results: Algorithm performance showed an area under the curve of 0.960 at the image patch level. More AI-assisted reviews identified AFBs than manual microscopy or WSI examination (P < .001). Sensitivity, negative predictive value, and accuracy were highest for AI-assisted reviews. AI-assisted reviews also had the highest rate of matching the original sign-out diagnosis, were less time-consuming, and were much easier for pathologists to perform (P < .001).

Conclusions: This study reports the successful development and clinical validation of an AI-based digital pathology system to screen for AFBs in anatomic pathology material. AI assistance proved to be more sensitive and accurate, took pathologists less time to screen cases, and was easier to use than either manual microscopy or viewing WSIs.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1093/ajcp/aqaa215DOI Listing
February 2021

Cytologic and histological features of rare nonepithelial and nonlymphoid tumors of the thyroid.

Cancer Cytopathol 2021 Jan 25. Epub 2021 Jan 25.

Department of Pathology, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts.

Thyroid tumors can be classified into epithelial, nonepithelial, and nonprimary lesions. Nonepithelial thyroid tumors are rare. They can be of primary origin within the thyroid gland, arise secondary to contiguous growth from adjacent tissues, or represent metastatic disease. The incidence of nonepithelial tumors of the thyroid is only 1% to 2%, most of which are lymphomas; the remainder includes mesenchymal and histiocytic tumors. This review examines the cytohistological features of various nonepithelial and nonlymphoid tumors of the thyroid, including vascular lesions, neural tumors (including granular cell tumor and paraganglioma), smooth muscle tumors, solitary fibrous tumor, histiocytic neoplasms (eg, Langerhans cell histiocytosis and Rosai-Dorfman disease), and follicular dendritic cell sarcoma. Their differential diagnosis is discussed, including recommendations to prevent the pitfall of mistaking these rare tumors for more common epithelial thyroid neoplasms.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1002/cncy.22404DOI Listing
January 2021

Postmortem Findings Associated With SARS-CoV-2: Systematic Review and Meta-analysis.

Am J Surg Pathol 2021 01 20. Epub 2021 Jan 20.

Department of Pathology, University of Pittsburgh Medical Center, Pittsburgh, PA Department of Pathology, Icahn School of Medicine at Mount Sinai, New York, NY Department of Pathology, University of Michigan, Ann Arbor, MI Department of Pathology and Diagnostics, University and Hospital Trust of Verona, Verona Department of Pathology, Cannizzaro Hospital, Catania, Italy Department of Pathology, University of Washington Medical Center, Seattle, WA.

Coronavirus Disease 2019 (COVID-19), caused by the novel Severe Acute Respiratory Syndrome-associated Coronavirus 2 (SARS-CoV-2), has become a global threat to public health. COVID-19 is more pathogenic and infectious than the prior 2002 pandemic caused by SARS-CoV-1. The pathogenesis of certain disease manifestations in COVID-19 such as diffuse alveolar damage (DAD) are thought to be similar to SARS-CoV-1. However, the exact pathogenesis of COVID-19 related deaths remains poorly understood. The aim of this article was to systematically summarize the rapidly emerging literature regarding COVID-19 autopsies. A meta-analysis was also conducted based on data accrued from preprint and published articles on COVID-19 (n=241 patients) and the results compared with postmortem findings associated with SARS-CoV-1 deaths (n=91 patients). Both autopsy groups included mostly adults of median age 70 years with COVID-19 and 50 years with SARS-CoV-1. Overall, prevalence of DAD was more common in SARS-CoV-1 (100.0%) than COVID-19 (80.9%) autopsies (P=0.001). Extrapulmonary findings among both groups were not statistically significant except for hepatic necrosis (P <0.001), splenic necrosis (P<0.006) and white pulp depletion (P <0.001) that were more common with SARS-CoV-1. Remarkable postmortem findings in association with COVID-19 apart from DAD include pulmonary hemorrhage, viral cytopathic effect within pneumocytes, thromboembolism, brain infarction, endotheliitis, acute renal tubular damage, white pulp depletion of the spleen, cardiac myocyte necrosis, megakaryocyte recruitment, and hemophagocytosis.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1097/PAS.0000000000001650DOI Listing
January 2021

Diagnostic mesothelioma biomarkers in effusion cytology.

Cancer Cytopathol 2021 Jan 19. Epub 2021 Jan 19.

Department of Pathology, University of Michigan, Ann Arbor, Michigan.

Malignant mesothelioma is a rare malignancy with a poor prognosis whose development is related to asbestos fiber exposure. An increasing role of genetic predisposition has been recognized recently. Pleural biopsy is the gold standard for diagnosis, in which the identification of pleural invasion by atypical mesothelial cell is a major criterion. Pleural effusion is usually the first sign of disease; therefore, a cytological specimen is often the initial or the only specimen available for diagnosis. Given that reactive mesothelial cells may show marked atypia, the diagnosis of mesothelioma on cytomorphology alone is challenging. Accordingly, cell block preparation is encouraged, as it permits immunohistochemical staining. Traditional markers of mesothelioma such as glucose transporter 1 (GLUT1) and insulin-like growth factor 2 mRNA-binding protein 3 (IMP3) are informative, but difficult to interpret when reactive proliferations aberrantly stain positive. BRCA1-associated protein 1 (BAP1) nuclear staining loss is highly specific for mesothelioma, but sensitivity is low in sarcomatoid tumors. Cyclin-dependent kinase inhibitor 2A (CDKN2A)/p16 homozygous deletion, assessed by fluorescence in situ hybridization, is more specific for mesothelioma with better sensitivity, even in the sarcomatoid variant. The surrogate marker methylthioadenosine phosphorylase (MTAP) has been found to demonstrate excellent diagnostic correlation with p16. The purpose of this review is to provide an essential appraisal of the literature regarding the diagnostic value of many of these emerging biomarkers for malignant mesothelioma in effusion cytology.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1002/cncy.22398DOI Listing
January 2021

HLA-G expression in melanomas.

Int Rev Immunol 2021 Jan 11:1-21. Epub 2021 Jan 11.

Department of Pathology and Diagnostics, University and Hospital Trust of Verona, Verona, Italy.

Objective: Human leukocyte antigen G (HLA-G) is a non-classical HLA class I molecule involved in inducing tolerance at the feto-maternal interface and in escape of immune response by tumor cells. The aim of the study is to review the published literature on the expression of HLA-G in malignant melanomas and its clinicopathological and prognostic correlates.

Methods: A systematic search was carried out in electronic databases. Studies dealing with HLA-G expression in surgically-removed human samples were retrieved and analyzed.

Results: Of 1737 retrieved articles, 16 were included. The main themes regarded HLA-G expression in malignant melanocytic lesions, assessed by immunohistochemistry (IHC), soluble or molecular techniques, and its relationship with clinicopathological features, such as tumor thickness and malignant behavior. Overall significant HLA-G expression was found in 460/843 tumors (55%), and specifically in 251/556 melanomas (45%) evaluated with IHC, in 208/250 cases (83%) examined with soluble methods and in 13/23 melanoma lesions (57%) tested with polymerase chain reaction. Despite the correlation with parameters indicating an aggressive behavior, no studies demonstrated any prognostic value of HLA-G expression. Furthermore, uveal melanomas were constantly negative for this biomarker.

Conclusion: Overall, published data indicate that while HLA-G is involved in the interactions between melanomas and the immune system, it is unlikely to be the only factor to play such a role, therefore making it difficult to designate it as a prognostically relevant molecule. Evidence further suggests that HLA-G is not implicated in the immunobiology of uveal melanomas.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1080/08830185.2020.1869732DOI Listing
January 2021

(Re) Defining the High-Power Field for Digital Pathology.

J Pathol Inform 2020 9;11:33. Epub 2020 Oct 9.

Department of Pathology, Memorial Sloan Kettering Cancer Center, New York, NY, USA.

Background: The microscope high-power field (HPF) is the cornerstone for histopathology diagnostic evaluation such as the quantification of mitotic figures, lymphocytes, and tumor grading. With traditional light microscopy, HPFs are typically evaluated by quantifying histologic events in 10 fields of view at × 400 magnification. In the era of digital pathology, new variables are introduced that may affect HPF evaluation. The aim of this study was to determine the parameters that influence HPF in whole slide images (WSIs).

Materials And Methods: Glass slides scanned on various devices (Leica's Aperio GT450, AT2, and ScanScope XT; Philips UltraFast Scanner; Hamamatsu's Nanozoomer 2.0HT; and 3DHistech's P1000) were compared to acquired digital slides reviewed on each vendor's respective WSI viewer software (e.g., Aperio ImageScope, ImageScope DX, Philips IMS, 3DHistech CaseViewer, and Hamamatsu NDP.view) and an in-house developed vendor-agnostic viewer. WSIs were reviewed at "×40" equivalent HPF on different sized monitors with varying display resolutions (1900 × 1080-4500 × 3000) and aspect ratios (e.g., Food and Drug Administration [FDA]-cleared 27" Philips PS27QHDCR, FDA-cleared 24" Dell MR2416, 24" Hewlett Packard Z24n G2, and 28" Microsoft Surface Studio). Digital and microscopic HPF areas were calculated and compared.

Results: A significant variation of HPF area occurred between differing monitor size and display resolutions with minor differences between WSI viewers. No differences were identified by scanner or WSIs scanned at different resolutions (e.g., 0.5, 0.25, 0.24, and 0.12 μm/pixel).

Conclusion: Glass slide HPF at × 400 magnification with conventional light microscopy was not equivalent to "×40" digital HPF areas. Digital HPF quantification may vary due to differences in the tissue area displayed by monitor sizes, display resolutions, and WSI viewers but not by scanner or scanning resolution. These findings will need to be further clinically validated with potentially new digital metrics for evaluation.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.4103/jpi.jpi_48_20DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7737490PMC
October 2020

An artificial intelligence algorithm for prostate cancer diagnosis in whole slide images of core needle biopsies: a blinded clinical validation and deployment study.

Lancet Digit Health 2020 08;2(8):e407-e416

Department of Pathology, University of Pittsburgh Medical Center, Pittsburgh, PA, USA.

Background: There is high demand to develop computer-assisted diagnostic tools to evaluate prostate core needle biopsies (CNBs), but little clinical validation and a lack of clinical deployment of such tools. We report here on a blinded clinical validation study and deployment of an artificial intelligence (AI)-based algorithm in a pathology laboratory for routine clinical use to aid prostate diagnosis.

Methods: An AI-based algorithm was developed using haematoxylin and eosin (H&E)-stained slides of prostate CNBs digitised with a Philips scanner, which were divided into training (1 357 480 image patches from 549 H&E-stained slides) and internal test (2501 H&E-stained slides) datasets. The algorithm provided slide-level scores for probability of cancer, Gleason score 7-10 (vs Gleason score 6 or atypical small acinar proliferation [ASAP]), Gleason pattern 5, and perineural invasion and calculation of cancer percentage present in CNB material. The algorithm was subsequently validated on an external dataset of 100 consecutive cases (1627 H&E-stained slides) digitised on an Aperio AT2 scanner. In addition, the AI tool was implemented in a pathology laboratory within routine clinical workflow as a second read system to review all prostate CNBs. Algorithm performance was assessed with area under the receiver operating characteristic curve (AUC), specificity, and sensitivity, as well as Pearson's correlation coefficient (Pearson's r) for cancer percentage.

Findings: The algorithm achieved an AUC of 0·997 (95% CI 0·995 to 0·998) for cancer detection in the internal test set and 0·991 (0·979 to 1·00) in the external validation set. The AUC for distinguishing between a low-grade (Gleason score 6 or ASAP) and high-grade (Gleason score 7-10) cancer diagnosis was 0·941 (0·905 to 0·977) and the AUC for detecting Gleason pattern 5 was 0·971 (0·943 to 0·998) in the external validation set. Cancer percentage calculated by pathologists and the algorithm showed good agreement (r=0·882, 95% CI 0·834 to 0·915; p<0·0001) with a mean bias of -4·14% (-6·36 to -1·91). The algorithm achieved an AUC of 0·957 (0·930 to 0·985) for perineural invasion. In routine practice, the algorithm was used to assess 11 429 H&E-stained slides pertaining to 941 cases leading to 90 Gleason score 7-10 alerts and 560 cancer alerts. 51 (9%) cancer alerts led to additional cuts or stains being ordered, two (4%) of which led to a third opinion request. We report on the first case of missed cancer that was detected by the algorithm.

Interpretation: This study reports the successful development, external clinical validation, and deployment in clinical practice of an AI-based algorithm to accurately detect, grade, and evaluate clinically relevant findings in digitised slides of prostate CNBs.

Funding: Ibex Medical Analytics.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/S2589-7500(20)30159-XDOI Listing
August 2020

The histopathological diagnosis of atypical meningioma: glass slide versus whole slide imaging for grading assessment.

Virchows Arch 2020 Dec 10. Epub 2020 Dec 10.

Department of Diagnostics and Public Health, Section of Pathology, University of Verona, Verona, Italy.

Limited studies on whole slide imaging (WSI) in surgical neuropathology reported a perceived limitation in the recognition of mitoses. This study analyzed and compared the inter- and intra-observer concordance for atypical meningioma, using glass slides and WSI. Two neuropathologists and two residents assessed the histopathological features of 35 meningiomas-originally diagnosed as atypical-in a representative glass slide and corresponding WSI. For each histological parameter and final diagnosis, we calculated the inter- and intra-observer concordance in the two viewing modes and the predictive accuracy on recurrence. The concordance rates for atypical meningioma on glass slides and on WSI were 54% and 60% among four observers and 63% and 74% between two neuropathologists. The inter-observer agreement was higher using WSI than with glass slides for all parameters, with the exception of high mitotic index. For all histological features, we found median intra-observer concordance of ≥ 79% and similar predictive accuracy for recurrence between the two viewing modes. The higher concordance for atypical meningioma using WSI than with glass slides and the similar predictive accuracy for recurrence in the two modalities suggest that atypical meningioma may be safely diagnosed using WSI.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1007/s00428-020-02988-1DOI Listing
December 2020

Challenges in the Development, Deployment, and Regulation of Artificial Intelligence in Anatomic Pathology.

Am J Pathol 2020 Nov 24. Epub 2020 Nov 24.

Department of Pathology, University of Michigan, Ann Arbor, Michigan.

Significant advances in artificial intelligence (AI), deep learning, and other machine-learning approaches have been made in recent years, with applications found in almost every industry, including health care. AI has proved to be capable of completing a spectrum of mundane to complex medically oriented tasks previously performed only by boarded physicians, most recently assisting with the detection of cancers difficult to find on histopathology slides. Although computers will not replace pathologists any time soon, properly designed AI-based tools hold great potential for increasing workflow efficiency and diagnostic accuracy in the practice of pathology. Recent trends, such as data augmentation, crowdsourcing for generating annotated data sets, and unsupervised learning with molecular and/or clinical outcomes versus human diagnoses as a source of ground truth, are eliminating the direct role of pathologists in algorithm development. Proper integration of AI-based systems into anatomic-pathology practice will necessarily require fully digital imaging platforms, an overhaul of legacy information-technology infrastructures, modification of laboratory/pathologist workflows, appropriate reimbursement/cost-offsetting models, and ultimately, the active participation of pathologists to encourage buy-in and oversight. Regulations tailored to the nature and limitations of AI are currently in development and, when instituted, are expected to promote safe and effective use. This review addresses the challenges in AI development, deployment, and regulation to be overcome prior to its widespread adoption in anatomic pathology.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.ajpath.2020.10.018DOI Listing
November 2020

Thyroid paraganglioma: A diagnostic pitfall in thyroid FNA.

Cancer Cytopathol 2020 Nov 24. Epub 2020 Nov 24.

Department of Pathology, University of Pittsburgh, Pittsburgh, Pennsylvania.

Background: Thyroid paragangliomas are extremely rare and often are misdiagnosed by preoperative fine-needle aspiration (FNA) because their cytologic features overlap with those of other thyroid neoplasms. The objective of this study was to review the cytomorphology in a series of thyroid paragangliomas and correlate the findings with histopathology.

Methods: Five thyroid paraganglioma cases that underwent FNA were reviewed. Their clinical presentation, radiology features, cytomorphology, ancillary tests, and histopathology were analyzed.

Results: All patients were women with an average age of 49 years (age range, 35-61 years) and presented with an asymptomatic, solitary thyroid nodule. Radiologically, these nodules (size range, 1.8-3.0 cm) were well circumscribed, hypoechoic, and hypervascular. FNA smears showed clusters of loosely cohesive, medium-to-large epithelioid cells with clear-to-eosinophilic and occasionally foamy cytoplasm that had indistinct cytoplasmic borders. The nuclei were round to oval with focal nuclear membrane irregularities, inconspicuous nucleoli, focal marked anisonucleosis, and occasional intranuclear pseudoinclusions. Naked nuclei, variable numbers of plasmacytoid cells, multinucleated giant cells, and sustentacular cells were present in the background along with blood vessels and lymphocytes. Cytology diagnoses were incorrect and included follicular neoplasm (n = 4) and follicular lesion of undetermined significance (n = 1). Final histopathology with immunohistochemistry revealed conventional paraganglioma (n = 3) or sclerosing paraganglioma with invasive features (n = 2).

Conclusions: All thyroid paragangliomas were misdiagnosed on FNA as follicular neoplasms, in part because of the rarity of these tumors in this location and cytomorphology mimicking follicles. The absence of colloid, the presence of naked nuclei, focal marked anisonucleosis, and the presence of sustentacular cells are important cytology clues.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1002/cncy.22390DOI Listing
November 2020

Fine needle aspiration of salivary gland carcinomas with high-grade transformation: A multi-institutional study of 22 cases and review of the literature.

Cancer Cytopathol 2020 Nov 19. Epub 2020 Nov 19.

Departments of Pathology, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts.

Background: High-grade transformation (HGT) is a rare process whereby conventional low- to intermediate-grade salivary gland carcinomas (SGC) transform into high-grade, poorly or undifferentiated malignancies with focal or complete loss of their conventional histomorphologic features. Because tumors with HGT are associated with a worse prognosis than their conventional counterparts, preoperative recognition of HGT may be of benefit for optimal patient management. Using a multi-institutional approach, we describe the largest fine needle aspiration (FNA) cohort of salivary gland carcinomas with HGT.

Methods: The archives of 9 large academic medical centers were searched, and 22 cases of SGC with HGT were identified by surgical excision accompanied by preoperative FNA. Clinical and cytomorphologic features were retrospectively reviewed.

Results: The male-to-female ratio was 14:8, and the mean patient age was 60.2 years. The average tumor size was 3.6 cm, and 19 cases were from the parotid gland. Acinic cell carcinoma with HGT was the most common tumor subtype, comprising 12 cases with HGT, followed by adenoid cystic carcinoma, secretory carcinoma, and other subtypes. Eighteen cases were classified as malignant; however, a specific diagnosis of HGT was not made. Sixteen cases contained a high-grade cytologic component, and 7 cases had a mixture of both conventional and high-grade components retrospectively.

Conclusions: SGC with HGT should be considered in the differential diagnosis of a salivary gland aspirate exhibiting high-grade cytomorphologic features. The presence of distinct tumor populations, conventional and high-grade, should prompt consideration of HGT, especially when the conventional component is acinic cell carcinoma or adenoid cystic carcinoma.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1002/cncy.22388DOI Listing
November 2020

Challenges Developing Deep Learning Algorithms in Cytology.

Acta Cytol 2020 Nov 2:1-9. Epub 2020 Nov 2.

Division of Anatomical Pathology, School of Pathology, University of the Witwatersrand and National Health Laboratory Service, Johannesburg, South Africa.

Background: The incorporation of digital pathology into routine pathology practice is becoming more widespread. Definite advantages exist with respect to the implementation of artificial intelligence (AI) and deep learning in pathology, including cytopathology. However, there are also unique challenges in this regard.

Summary: This review discusses cytology-specific challenges, including the need to implement digital cytology prior to AI; the large file sizes and increased acquisition times for whole slide images in cytology; the routine use of multiple stains, such as Papanicolaou and Romanowsky stains; the lack of high-quality annotated datasets on which to train algorithms; and the considerable computer resources required, in terms of both computer infrastructure and skilled personnel, for computing and storage of data. Global concerns regarding AI that are certainly applicable to cytology include the need for model validation and continued quality assurance, ethical issues such as the use of patient data in developing algorithms, the need to develop regulatory frameworks regarding what type of data can be utilized and ensuring cybersecurity during data collection and storage, and algorithm development. Key Messages: While AI will likely play a role in cytology practice in the future, applying this technology to cytology poses a unique set of challenges. A broad understanding of digital pathology and algorithm development is desirable to guide the development of algorithms, as well as the need to be cognizant of potential pitfalls to avoid when incorporating the technology in practice.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1159/000510991DOI Listing
November 2020

Is the Time Right to Start Using Digital Pathology and Artificial Intelligence for the Diagnosis of Lymphoma?

J Pathol Inform 2020 26;11:16. Epub 2020 Jun 26.

University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, USA.

View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.4103/jpi.jpi_16_20DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7513776PMC
June 2020

Application of the Milan System for Reporting Salivary Gland Cytopathology to cystic salivary gland lesions.

Cancer Cytopathol 2020 Oct 1. Epub 2020 Oct 1.

Department of Pathology, University of Michigan, Ann Arbor, Michigan.

Background: Cystic salivary gland lesions present diagnostic challenges on fine-needle aspiration (FNA) specimens that are related to sampling limitations and a broad differential diagnosis. This study evaluated the benefit of applying the Milan System for Reporting Salivary Gland Cytopathology (MSRSGC) to a series of cystic salivary gland lesions.

Methods: The pathology archives at the Johns Hopkins Hospital were searched to identify cystic salivary gland FNA specimens over a 19-year period (2000-2018). Patient demographics, cytomorphologic features, and clinical and surgical follow-up were recorded. The MSRSGC was applied to the cases. The risk of malignancy (ROM) and the risk of neoplasia (RON) were calculated for each category.

Results: One hundred seventy-eight cases were identified (96 males and 82 females) with a mean age of 53 years (range, 4-90 years). After the MSRSGC was applied, there were 52 nondiagnostic cases (29.2%), 80 nonneoplastic cases (44.9%), 35 cases of atypia of undetermined significance (AUS; 19.7%), 3 benign neoplasms (1.7%), 3 salivary gland neoplasms of uncertain malignant potential (SUMP; 1.7%), 4 cases suspicious for malignancy (SFM; 2.2%), and 1 malignant case (0.6%). One hundred fifty-six of the 178 patients (87.6%) had follow-up data available. The RON and ROM values for cases with surgical follow-up were 33.3% (3 of 9) and 22.2% (2 of 9) for the nondiagnostic category, 42.9% (9 of 21) and 19% (4 of 21) for the nonneoplastic category, 76.5% (13 of 17) and 29.4% (5 of 17) for the AUS category, 100.0% (2 of 2) and 50.0% (1 of 2) for the SUMP category, and 100% (2 of 2) and 100% (2 of 2) for the SFM category, respectively.

Conclusions: Applying the MSRSGC to cystic salivary gland lesions improves patient management by preventing unnecessary surgery for nonneoplastic conditions. The ROM was highest in the SFM category (100%), which was followed by the SUMP, AUS, nondiagnostic, and nonneoplastic categories. Less than adequate specimens may increase the diagnosis of AUS.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1002/cncy.22363DOI Listing
October 2020

Corpora amylacea in sputum smears: Incidence and clinical significance.

Cytopathology 2021 01 25;32(1):108-114. Epub 2020 Oct 25.

Department of Pathology & Clinical Labs, University of Michigan, Ann Arbor, MI, USA.

Objective: To examine conventional sputum smears for the presence of corpora amylacea (CA), determining their incidence and clinical significance.

Methods: A retrospective 4-year cohort study was undertaken of sputum samples from 1176 consecutive patients for the presence of CA. Variables such as age, sex, smoking status, and the presence or absence of haemoptysis were extracted from the medical record. A random group of 50 patients was selected as a control group, and a random group of 50 patients whose ages were below 49 years was also included as an age-based control.

Results: A total of 1075 of the initial cohort of consecutive patients were included in the study. from these, there were 6898 sputum smears, of which 1.91% (132 smears) contained CA, corresponding to 9.86% of the cohort of patients (106 patients). There was a strong, positive, statistically significant correlation between age and CA presence (τ  = .402, P < .001), which supports that CA are associated with older patients. The results of a binary logistic regression indicated that there was a significant association between age, diagnosis of chronic obstructive pulmonary disease and CA presence (χ  = 49.051, df = 2, P < .001).

Conclusions: The presence of CA in sputum smears is related to age, being much more frequent in older people. Moreover, CA are related to non-neoplastic lung diseases.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1111/cyt.12919DOI Listing
January 2021

Cytomorphology and diagnostic pitfalls of sebaceous and nonsebaceous salivary gland lymphadenoma: A multi-institutional study.

Diagn Cytopathol 2021 Jan 14;49(1):83-95. Epub 2020 Sep 14.

Department of Pathology, Massachusetts General Hospital, Boston, Massachusetts, USA.

Background: Salivary gland lymphadenoma (LAD) is a rare benign neoplasm comprising sebaceous (SLAD) and nonsebaceous (NSLAD) types. Despite established histologic criteria, limited data on cytomorphology, tumor heterogeneity, and overlap with other entities make the diagnosis of LAD by fine needle aspiration (FNA) challenging. We describe a multi-institutional cohort of 14 LADs with cytology, clinical, radiologic, and histopathologic data.

Methods: Our cohort included nine SLAD and five NSLAD with corresponding histopathology. Mean patient age and M:F ratio were 60.4 years (range 45-86 years) and 1:2 for SLADs and 57.4 years (range 42-80 years) and 1:1.5 for NSLADs, respectively. One NSLAD patient had a germline predisposition for Cowden syndrome. Glass slides and whole slide images of air-dried Diff-Quik (DQ), alcohol-stained Papanicolaou smears (Pap) and cellblocks were reviewed for key cytomorphologic findings.

Results: FNAs from SLAD and NSLADs demonstrated vacuolated and basaloid epithelial clusters within a lymphoid background. Vacuolated cells from SLAD showed sebaceous cells with microvesicular cytoplasm indenting a central nucleus. Vacuolated cells from NSLAD were columnar with eccentric nuclei, corresponding to abluminal glandular cells. SLADs were classified using the Milan System for Reporting Salivary Gland Cytopathology as nondiagnostic (11.1%), nonneoplastic (44.4%), atypia of uncertain significance (AUS) (22.2%), and salivary gland neoplasm of uncertain malignant potential (SUMP) (22.2%). NSLADs were classified as AUS (40%), SUMP (40%) and Benign Neoplasm (20%).

Conclusion: Although rare, knowing the cytologic features of salivary LAD is important to avoid diagnostic pitfalls. Vacuolated cells can be prominent in both SLAD and NSLAD aspirates. Diagnostic issues arise from insufficient sampling of all tumor components leading to marked variation in diagnostic classification of LAD.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1002/dc.24602DOI Listing
January 2021

A Digital Pathology Solution to Resolve the Tissue Floater Conundrum.

Arch Pathol Lab Med 2021 Mar;145(3):359-364

Kimia Lab, University of Waterloo, Waterloo, Ontario, Canada (Kalra, Babaie, Tizhoosh).

Context.—: Pathologists may encounter extraneous pieces of tissue (tissue floaters) on glass slides because of specimen cross-contamination. Troubleshooting this problem, including performing molecular tests for tissue identification if available, is time consuming and often does not satisfactorily resolve the problem.

Objective.—: To demonstrate the feasibility of using an image search tool to resolve the tissue floater conundrum.

Design.—: A glass slide was produced containing 2 separate hematoxylin and eosin (H&E)-stained tissue floaters. This fabricated slide was digitized along with the 2 slides containing the original tumors used to create these floaters. These slides were then embedded into a dataset of 2325 whole slide images comprising a wide variety of H&E stained diagnostic entities. Digital slides were broken up into patches and the patch features converted into barcodes for indexing and easy retrieval. A deep learning-based image search tool was employed to extract features from patches via barcodes, hence enabling image matching to each tissue floater.

Results.—: There was a very high likelihood of finding a correct tumor match for the queried tissue floater when searching the digital database. Search results repeatedly yielded a correct match within the top 3 retrieved images. The retrieval accuracy improved when greater proportions of the floater were selected. The time to run a search was completed within several milliseconds.

Conclusions.—: Using an image search tool offers pathologists an additional method to rapidly resolve the tissue floater conundrum, especially for those laboratories that have transitioned to going fully digital for primary diagnosis.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.5858/arpa.2020-0034-OADOI Listing
March 2021

Digital cytology: Look how much has been achieved.

Cytopathology 2020 09;31(5):370-371

Department of Pathology, University of Pittsburgh Medical Center, Pittsburgh, PA, USA.

View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1111/cyt.12866DOI Listing
September 2020

Hematoidin crystals in sputum smears: Cytopathology and clinical associations.

Ann Thorac Med 2020 Jul-Sep;15(3):155-162. Epub 2020 Jun 18.

Department of Pathology, University of Pittsburgh Medical Center Shadyside, Pittsburgh, Pennsylvania, USA.

Background: There has been limited correlation of hematoidin crystals (HC) in sputum with clinical and diagnostic characteristics, partly because they are difficult to recognize. The aim of this study was, therefore, to assess these relationships in a consecutive cohort of sputum samples from patients submitted to a cytology laboratory.

Methods: Adequate sputum samples from 489 individuals were enrolled in this study. These were fixed in ethanol, stained by the Papanicolaou method and examined microscopically. The normality of the distribution of the continuous variable (age in years) was examined using the Shapiro-Wilk normality test. As the data were not normally distributed, Kendall rank correlation was used to correlate age with the presence of HC. The Pearson's Chi-square test was used to determine if the proportion of cases with the presence of HC was different among the categorical data variables. A univariate binary logistic regression was used to determine the variables most strongly associated with HC presence. The results include odds ratios, 95% confidence intervals, Wald χ statistics, and corresponding values, with statistical significance assumed at ≤ 0.05. Analyses were performed using IBM SPSS Statistics 22.

Results: The presence of HC in sputum was associated with increasing age ( < 0.01), current smoking ( < 0.001), chronic occupational exposure to dust ( = 0.001), and hemoptysis ( < 0.001). These crystals were most prevalent in patients with a diagnosis of carcinoma (93.9%), bronchiectasis (48%), silicosis (16.0%), and chronic obstructive pulmonary disease (0.8%). Crystals were not found in patients with acute bronchitis, asthma, or lung infections, including viral and bacterial pneumonia.

Conclusions: There are clear associations linked to the finding of HC in sputum that may be helpful in pointing to a specific diagnosis, such as the possibility of underlying carcinoma.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.4103/atm.ATM_69_20DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7423204PMC
June 2020

Foreword: JASC small biopsy special edition.

J Am Soc Cytopathol 2020 Sep - Oct;9(5):305

Department of Pathology, University of Michigan, Ann Arbor, MI. Electronic address:

View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.jasc.2020.07.001DOI Listing
August 2020

Computational Cytology: Lessons Learned from Pap Test Computer-Assisted Screening.

Acta Cytol 2020 Jul 21:1-15. Epub 2020 Jul 21.

Department of Pathology, University of Pittsburg Medical Center, Pittsburgh, Pennsylvania, USA.

Background: In the face of rapid technological advances in computational cytology including artificial intelligence (AI), optimization of its application to clinical practice would benefit from reflection on the lessons learned from the decades-long journey in the development of computer-assisted Pap test screening.

Summary: The initial driving force for automated screening in cytology was the overwhelming number of Pap tests requiring manual screening, leading to workflow backlogs and incorrect diagnoses. Several companies invested resources to address these concerns utilizing different specimen processing techniques and imaging systems. However, not all companies were commercially prosperous. Successful implementation of this new technology required viable use cases, improved clinical outcomes, and an acceptable means of integration into the daily workflow of cytopathology laboratories. Several factors including supply and demand, Food and Drug Administration (FDA) oversight, reimbursement, overcoming learning curves and workflow changes associated with the adoption of new technology, and cytologist apprehension, played a significant role in either promoting or preventing the widespread adoption of automated screening technologies. Key Messages: Any change in health care, particularly those involving new technology that impacts clinical workflow, is bound to have its successes and failures. However, perseverance through learning curves, optimizing workflow processes, improvements in diagnostic accuracy, and regulatory and financial approval can facilitate widespread adoption of these technologies. Given their history with successfully implementing automated Pap test screening, cytologists are uniquely positioned to not only help with the development of AI technology for other areas of pathology, but also to guide how they are utilized, regulated, and managed.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1159/000508629DOI Listing
July 2020

The Future of Pathology: What can we Learn from the COVID-19 Pandemic?

J Pathol Inform 2020 9;11:15. Epub 2020 Jun 9.

Department of Histopathology, Leeds Teaching Hospitals NHS Trust, Leeds, UK.

View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.4103/jpi.jpi_29_20DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7363060PMC
June 2020

Decidual Vasculopathy Identification in Whole Slide Images Using Multiresolution Hierarchical Convolutional Neural Networks.

Am J Pathol 2020 10 15;190(10):2111-2122. Epub 2020 Jul 15.

Department of Mechanical Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania. Electronic address:

After a child is born, the examination of the placenta by a pathologist for abnormalities, such as infection or maternal vascular malperfusion, can provide important information about the immediate and long-term health of the infant. Detection of the pathologic placental blood vessel lesion decidual vasculopathy (DV) has been shown to predict adverse pregnancy outcomes, such as preeclampsia, which can lead to mother and neonatal morbidity in subsequent pregnancies. However, because of the high volume of deliveries at large hospitals and limited resources, currently a large proportion of delivered placentas are discarded without inspection. Furthermore, the correct diagnosis of DV often requires the expertise of an experienced perinatal pathologist. We introduce a hierarchical machine learning approach for the automated detection and classification of DV lesions in digitized placenta slides, along with a method of coupling learned image features with patient metadata to predict the presence of DV. Ultimately, the approach will allow many more placentas to be screened in a more standardized manner, providing feedback about which cases would benefit most from more in-depth pathologic inspection. Such computer-assisted examination of human placentas will enable real-time adjustment to infant and maternal care and possible chemoprevention (eg, aspirin therapy) to prevent preeclampsia, a disease that affects 2% to 8% of pregnancies worldwide, in women identified to be at risk with future pregnancies.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.ajpath.2020.06.014DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7668102PMC
October 2020

Is thyroid core needle biopsy a valid compliment to fine-needle aspiration?

J Am Soc Cytopathol 2020 Sep - Oct;9(5):383-388. Epub 2020 Jun 18.

Division of Anatomic Pathology and Histology, Fondazione Policlinico "Agostino Gemelli"-IRCCS, Rome, Italy. Electronic address:

Fine-needle aspiration (FNA) has long been considered the first and an important diagnostic tool in the evaluation of thyroid nodules. The advantages of FNA include simplicity, safety, cost-effectiveness, high diagnostic accuracy, and low complication rate. Nevertheless, limitations associated with FNA include a substantial rate of inconclusive results and indeterminate interpretations. Therefore, core needle biopsy (CNB) of the thyroid gland has been proposed as a complementary or even alternate diagnostic method to evaluate thyroid nodules. Although controversial, a growing number of researchers have reported CNB to be an effective and safe sampling method for thyroid nodules, especially for cases with inadequate or indeterminate FNA yields. Skeptics highlight local pain and bleeding risk. Supporters highlight the potential likelihood of overcoming FNA limitations by obtaining a larger amount of tissue and using architecture and cellular details to guide possible ancillary testing. This review evaluates the indications, advantages, and disadvantages of CNB as compared with FNA of the thyroid gland.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.jasc.2020.06.003DOI Listing
June 2020

Intratumoral budding and automated CD8-positive T-cell density in pretreatment biopsies can predict response to neoadjuvant therapy in rectal adenocarcinoma.

Mod Pathol 2021 01 13;34(1):171-183. Epub 2020 Jul 13.

Department of Pathology, University of Pittsburgh Medical Center, Pittsburgh, PA, USA.

Tumor budding and CD8-positive (+) T-cells are recognized as prognostic factors in colorectal adenocarcinoma. We assessed CD8+ T-cell density and intratumoral budding in pretreatment rectal cancer biopsies to determine if they are predictive biomarkers for response to neoadjuvant therapy and survival. Pretreatment biopsies of locally advanced rectal adenocarcinoma from 117 patients were evaluated for CD8+ T-cell density using automated quantitative digital image analysis and for intratumoral budding and correlated with clinicopathological variables on postneoadjuvant surgical resection specimens, response to neoadjuvant therapy, and survival. Patients with high CD8+ T-cell density (≥157 per mm) on biopsy were significantly more likely to exhibit complete/near complete response to neoadjuvant therapy (66% vs. 33%, p = 0.001) and low tumor stage (0 or I) on resection (62% vs. 30%, p = 0.001) compared with patients with low CD8+ T-cell density. High CD8+ T-cell density was an independent predictor of response to neoadjuvant therapy with a 2.63 higher likelihood of complete response (95% CI 1.04-6.65, p = 0.04) and a 3.66 higher likelihood of complete/near complete response (95% CI 1.60-8.38, p = 0.002). The presence of intratumoral budding on biopsy was significantly associated with a reduced likelihood of achieving complete/near complete response to neoadjuvant therapy (odds ratio 0.36, 95% CI 0.13-0.97, p = 0.048). Patients with intratumoral budding on biopsy had a significantly reduced disease-free survival compared with patients without intratumoral budding (5-year survival 39% vs 87%, p < 0.001). In the multivariable model, the presence of intratumoral budding on biopsy was associated with a 3.35-fold increased risk of tumor recurrence (95% CI 1.25-8.99, p = 0.02). In conclusion, CD8+ T-cell density and intratumoral budding in pretreatment biopsies of rectal adenocarcinoma are independent predictive biomarkers of response to neoadjuvant therapy and intratumoral budding associates with patient survival. These biomarkers may be helpful in selecting patients who will respond to neoadjuvant therapy and identifying patients at risk for recurrence.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/s41379-020-0619-8DOI Listing
January 2021

Relevance of rosette patterns in variants of papillary thyroid carcinoma.

Cytopathology 2020 11 10;31(6):533-540. Epub 2020 Aug 10.

Department of Anatomic Pathology and Histology-Fondazione policlinico universitario, "Agostino Gemelli"-IRCCS, Rome, Italy.

Introduction: The detection of rosette-like clusters (RLC) of follicular cells in thyroid carcinoma has been reported mostly in the columnar cell variant of papillary thyroid carcinoma (PTC). Despite the fact that diagnosing variants of PTC is no longer encouraged by The Bethesda System for Reporting Thyroid Cytopathology, the identification of cytomorphological features such as RLC linked with these tumours might help reduce possible misinterpretation in thyroid fine needle aspiration (FNA) cytology. We accordingly investigated the potential correlation of architectural patterns including RLC with PTC variants.

Methods: We analysed 225 thyroid FNA cytology cases diagnosed as suspicious for malignancy (SFM) and positive for malignancy (M) over a 1-year time where all samples had corresponding histology. We also included 150 benign lesions from the same period. The presence of RLC vs similar appearing solid clusters, papillary structures and microfollicles were evaluated. We also performed immunocytochemistry and molecular testing for BRAFV600E.

Results: We included 100 (44.4%) SFM favouring PTC and 125 (55.6%) M cases with cyto-histological correlation. On histology, all SFM and M cases showed malignancy including 140 (62.2%) classic PTC and 85 (37.8%) PTC variants. The cytomorphological patterns in all FNA samples included solid (74%), papillary (89%), microfollicular (70%), and pseudo-RLC morphology (25.7%). We identified only pseudo-RLC in 33 FNA specimens from PTC variant cases that included tall cell variant (42.4%), hobnail variant (21.2%) and miscellaneous variants (36.3%) of PTC. No definitive RLC were detected in our series. Immunocytochemistry and BRAFV600E were not specifically linked with an RLC pattern.

Conclusions: These findings demonstrate that in our dataset the architectural pattern of RLC was not recognised within PTC variants. However, we did identify a pseudo-RLC pattern that was observed in association with tall cell variant and hobnail variant cases of PTC.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1111/cyt.12885DOI Listing
November 2020

Small core needle biopsies in cytology practice: a survey of members of the American Society of Cytopathology.

J Am Soc Cytopathol 2020 Sep - Oct;9(5):310-321. Epub 2020 Jun 18.

Wisconsin State Laboratory of Hygiene, University of Wisconsin School of Medicine and Public Health, Department of Pathology and Laboratory Medicine, Madison, Wisconsin.

Introduction: The introduction of a new generation of core needle biopsies (CNBs) for endoscopic procedures has prompted reconsideration of the role of cytopathologists in the handling of small biopsies. The American Society of Cytopathology (ASC) has therefore conducted a survey with the intention of elucidating current practices regarding the handling of small CNBs.

Materials And Methods: The membership of the ASC was invited by email to participate in an online survey over a 2-month period. The survey consisted of 20 multiple choice questions with 2-8 possible responses per question.

Results: Of 2651 members contacted by e-mail, 282 (10.6%) responded to the survey questions, including 196 pathologists (69.5%) and 86 cytotechnologists (30.5%). Of these, 265 respondents were from the US/Canada (94.0%), with 156 from academic institutions (58.9%) and 109 from non-academic practices (41.1%); 17 were from other countries (6.0%). In 18.8% of all practices, cytopathologists sign out >90% of small CNBs from endoscopic and radiologically guided procedures; in 36.5% of practices >90% are signed out by surgical pathologists; the remainder have such cases divided more evenly between cytopathologists and surgical pathologists. Responses show that 78.0% of all respondents are interested in signing out more small biopsies in the future, and 80.5% desire increased small biopsy-related resources from the ASC.

Conclusions: The survey responses indicate that practices currently vary widely across institutions. Most indicated an interest in greater incorporation of small biopsies into the practice of cytopathology.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.jasc.2020.06.002DOI Listing
June 2020