-
Advances in Anatomic Pathology Nov 2020Artificial intelligence (AI) is having an increasing impact on the field of pathology, as computation techniques allow computers to perform tasks previously performed by... (Review)
Review
Artificial intelligence (AI) is having an increasing impact on the field of pathology, as computation techniques allow computers to perform tasks previously performed by people. Here, we offer a simple and practical guide to AI methods used in pathology, such as digital image analysis, next-generation sequencing, and natural language processing. We not only provide a comprehensive review, but also discuss relevant history and future directions of AI in pathology. We additionally provide a short tabular dictionary of AI terminology which will help practicing pathologists and researchers to understand this field.
Topics: Artificial Intelligence; High-Throughput Nucleotide Sequencing; Humans; Machine Learning; Pathology
PubMed: 32773432
DOI: 10.1097/PAP.0000000000000277 -
Alzheimer's & Dementia : the Journal of... Jul 2023Post-mortem analysis provides definitive diagnoses of neurodegenerative diseases; however, only a few can be diagnosed during life.
INTRODUCTION
Post-mortem analysis provides definitive diagnoses of neurodegenerative diseases; however, only a few can be diagnosed during life.
METHODS
This study employed statistical tools and machine learning to predict 17 neuropathologic lesions from a cohort of 6518 individuals using 381 clinical features (Table S1). The multisite data allowed validation of the model's robustness by splitting train/test sets by clinical sites. A similar study was performed for predicting Alzheimer's disease (AD) neuropathologic change without specific comorbidities.
RESULTS
Prediction results show high performance for certain lesions that match or exceed that of research annotation. Neurodegenerative comorbidities in addition to AD neuropathologic change resulted in compounded, but disproportionate, effects across cognitive domains as the comorbidity number increased.
DISCUSSION
Certain clinical features could be strongly associated with multiple neurodegenerative diseases, others were lesion-specific, and some were divergent between lesions. Our approach could benefit clinical research, and genetic and biomarker research by enriching cohorts for desired lesions.
Topics: Humans; Alzheimer Disease; Comorbidity; Neuropathology; Biomarkers
PubMed: 36681388
DOI: 10.1002/alz.12921 -
Ocular Immunology and Inflammation 2022To report the cytopathology of vitreous biopsy samples in patients with acute retinal necrosis (ARN) who underwent pars plana vitrectomy (PPV). We also describe two...
PURPOSE
To report the cytopathology of vitreous biopsy samples in patients with acute retinal necrosis (ARN) who underwent pars plana vitrectomy (PPV). We also describe two patients with unique clinical courses, cytopathologic findings, and immune response.
METHODS
A retrospective review of patients with ARN who developed retinal detachment (RD) and underwent PPV from 22011 to 2019 at the Emory Eye Center was performed to assess cytopathology findings of vitreous biopsy samples. Patient demographics and laboratory testing including aqueous humor PCR for viral pathogens were recorded. Additional clinical details abstracted included intravitreal injections, surgical procedures, and vitreous cytopathological reports including immunohistochemistry findings.
RESULTS
Fourteen eyes of 12 patients with RD were reviewed. Ten eyes showed HSV DNA (71%) and 4 demonstrated VZV DNA (29%). All eyes received intravitreal antivirals (i.e. ganciclovir or foscarnet) with a median of 8.5 intravitreal injections per eye. Diagnoses prompting PPV included tractional RD in 14 eyes (100%), rhegmatogenous RD in 8 eyes (57%), vitreous hemorrhage in 4 eyes (29%) and vitreous opacity in 4 (29%). Ophthalmic pathology reports showed lymphocyte populations in 10 eyes (71%) with a CD3 + T-cell predominance in two patients where immunohistochemistry of CD3+ and CD20+ for T- and B-cell populations was performed. Observed immune cell populations included macrophages or histiocytes (11 eyes, 79%) and polymorphonuclear cells in 4 eyes (29%). Initial median VA was 2.5 (IQR 2.0-3.0) and improved to 2.0 (IQR 1.48-3.00, = .48) at 6-months and 1.8 (IQR 1.2-3.0, = .45) at 12 months follow-up.
CONCLUSIONS
Our cohort of ARN patients undergoing PPV show a spectrum of immunologic findings with the majority demonstrating a lymphocytic response. Histiocytes, macrophages, and PMNs were also observed. Cytopathologic and immunologic studies suggest that both innate and adaptive immunity are responsible for the clinical disease findings observed in ARN. The variability of the response to treatment in patients with ARN may reflect patient-to-patient differences in their antigen-specific immune response. Understanding the immunologic response associated with ARN may provide valuable information regarding the dosing and timing of treatment.
Topics: Humans; Retinal Necrosis Syndrome, Acute; Cytology
PubMed: 34242097
DOI: 10.1080/09273948.2021.1922926 -
The American Journal of Dermatopathology Mar 2018To review the Royal College of Pathologists of Australasia (RCPA) Quality Assurance Program Dermatopathology module from 2005 to 2016 to assess diagnostic performance,... (Review)
Review
AIMS
To review the Royal College of Pathologists of Australasia (RCPA) Quality Assurance Program Dermatopathology module from 2005 to 2016 to assess diagnostic performance, changes over time, and areas of diagnostic difficulty.
METHODS
The computerized records of the RCPA Dermatopathology subspecialist module were reviewed. Cases were categorized into groups including nonneoplastic disorders, neoplasms, and cases with multiple diagnoses. The performance of participants over time in each of these categories and in more specific areas (including melanocytic and adnexal neoplasms) was assessed. Cases which showed high rates of discordant responses were specifically reviewed.
RESULTS
One hundred sixteen cases circulated over 10 years were evaluated. The overall concordance rate was 77%, with a major discordance rate of 7%. There was a slightly higher concordance rate for neoplasms compared with nonneoplastic lesions (80% vs. 74%). Specific areas associated with lower concordance rates included classification of adnexal tumors and identification of multiple pathologies. A spindle cell nevus of Reed yielded a 40% discordance rate, with most misclassifications indicating melanoma.
CONCLUSIONS
The RCPA quality assurance program module has circulated a wide range of common and uncommon cases to participants over the 12 years studied, highlighting a low but important rate of major discordant responses. Melanocytic lesions, hematolymphoid infiltrates, adnexal tumors, and identification of multiple pathologies are identified as areas worthy of particular attention in quality improvement activities.
Topics: Dermatology; Humans; Observer Variation; Pathology; Quality Assurance, Health Care; Retrospective Studies; Skin Diseases
PubMed: 28953010
DOI: 10.1097/DAD.0000000000000922 -
Seminars in Nuclear Medicine Mar 2019Although electronic imaging was performed in the early 1950s in nuclear medicine, it was the introduction of computed tomography in 1972 that caused a revolution in... (Review)
Review
Although electronic imaging was performed in the early 1950s in nuclear medicine, it was the introduction of computed tomography in 1972 that caused a revolution in medical imaging in that it marked the beginning of the inevitable transformation to digital imaging. This transformation is now more or less complete. While initially these CT images were relatively small, comprised of only about 6400 pixels per slice, the steady move toward higher spatial resolution, multislice imaging, digital radiography, and fluoroscopy rapidly increased the size of images and the amount of data required to be stored, processed, displayed, and moved about in a medical imaging department. The more recent introduction of digital pathology with submicron-sized pixels and the need for color further increases these demands. Rising work volumes in hospital, a push for cost containment, and a move toward greater precision in diagnosis and treatment of disease all work together to motivate the development of automated image analysis algorithms and techniques to improve efficiencies in in vivo imaging and pathology. This may require bringing together information from different imaging and nonimaging sources within the institution. While technological development has provided practical means for storage of the burgeoning data load and the use of multiple processors and high-speed networks has enabled more sophisticated analysis locally or in the cloud, challenges remain in terms of the ability to integrate data from different systems, the development of appropriately annotated image bases for training and testing of algorithms, and issues around privacy and ownership in obtaining access to patient-related data.
Topics: Big Data; Diagnostic Imaging; Humans; Image Processing, Computer-Assisted; Mammography; Pathology; Precision Medicine
PubMed: 30819400
DOI: 10.1053/j.semnuclmed.2018.11.010 -
Journal of Medical Internet Research Apr 2023There is increasing interest in the use of artificial intelligence (AI) in pathology to increase accuracy and efficiency. To date, studies of clinicians' perceptions of... (Review)
Review
BACKGROUND
There is increasing interest in the use of artificial intelligence (AI) in pathology to increase accuracy and efficiency. To date, studies of clinicians' perceptions of AI have found only moderate acceptability, suggesting the need for further research regarding how to integrate it into clinical practice.
OBJECTIVE
The aim of the study was to determine contextual factors that may support or constrain the uptake of AI in pathology.
METHODS
To go beyond a simple listing of barriers and facilitators, we drew on the approach of realist evaluation and undertook a review of the literature to elicit stakeholders' theories of how, for whom, and in what circumstances AI can provide benefit in pathology. Searches were designed by an information specialist and peer-reviewed by a second information specialist. Searches were run on the arXiv.org repository, MEDLINE, and the Health Management Information Consortium, with additional searches undertaken on a range of websites to identify gray literature. In line with a realist approach, we also made use of relevant theory. Included documents were indexed in NVivo 12, using codes to capture different contexts, mechanisms, and outcomes that could affect the introduction of AI in pathology. Coded data were used to produce narrative summaries of each of the identified contexts, mechanisms, and outcomes, which were then translated into theories in the form of context-mechanism-outcome configurations.
RESULTS
A total of 101 relevant documents were identified. Our analysis indicates that the benefits that can be achieved will vary according to the size and nature of the pathology department's workload and the extent to which pathologists work collaboratively; the major perceived benefit for specialist centers is in reducing workload. For uptake of AI, pathologists' trust is essential. Existing theories suggest that if pathologists are able to "make sense" of AI, engage in the adoption process, receive support in adapting their work processes, and can identify potential benefits to its introduction, it is more likely to be accepted.
CONCLUSIONS
For uptake of AI in pathology, for all but the most simple quantitative tasks, measures will be required that either increase confidence in the system or provide users with an understanding of the performance of the system. For specialist centers, efforts should focus on reducing workload rather than increasing accuracy. Designers also need to give careful thought to usability and how AI is integrated into pathologists' workflow.
Topics: Humans; Artificial Intelligence; Narration; Machine Learning; Pathology
PubMed: 37093631
DOI: 10.2196/38039 -
International Journal of Surgical... Sep 2023Frozen sections (FS) are common in neurosurgery to address varied clinical concerns. Artifacts in central nervous system (CNS) FS can be severe and affect or hinder...
Frozen sections (FS) are common in neurosurgery to address varied clinical concerns. Artifacts in central nervous system (CNS) FS can be severe and affect or hinder interpretation. We performed a case-control study using a semiquantitative scale: the Histologic Preservation Score (HPS), and a quantitative scale: the Ice Crystal Vacuolization Score (ICVS), to compare the histologic quality yielded by snap- versus cryostat freezing techniques. All specimens were sectioned in 2 halves, one half was used for FS and the other for permanent evaluation. HPS assigns a distortion score to the FS sample using the non-frozen half as the comparator: 1 = minimal, 2 = slight, 3 = moderate, 4 & 5 = severe. The ICVS is the average size in µm of the 5 largest vacuoles/0.05 mm, evaluated on digitized slides. 86 CNS-FS were collected: 22 snap- and 64 cryostat-FS. Significant differences in HPS: 2.28 versus 2.84 () and ICVS 7.47 versus 14.56 ( ) were obtained for snap- versus cryostat-FS, respectively. HPS and ICVS showed a strong correlation: R = 0.63, Histologic distortion was worse for neuroglial than mesenchymal tissue by both methods; however, a significant difference was only observed in cryostat-FS: HPS: 3.23 versus 2.33, ; ICVS: 16.86 μm versus 10.26 μm, . Snap-FS yields better histologic quality than cryostat-FS for CNS-FS, and the difference is more pronounced in neuroglial samples. HPS and ICVS correlate strongly, indicating that the histologic quality is inversely proportional to water-crystallization. These results may apply to other areas of surgical pathology.
Topics: Humans; Case-Control Studies; Neurosurgical Procedures; Frozen Sections; Referral and Consultation; Pathology, Surgical
PubMed: 35971290
DOI: 10.1177/10668969221117987 -
American Journal of Clinical Pathology Nov 2021Consultation on surgical pathology specimens is part of the daily professional practice of every pathologist. We evaluated the characteristics of a good consultant and...
OBJECTIVES
Consultation on surgical pathology specimens is part of the daily professional practice of every pathologist. We evaluated the characteristics of a good consultant and the habits that should be avoided.
METHODS
A 1-page questionnaire was prepared to evaluate how pathologists select their consultants.
RESULTS
The questionnaire was emailed to 106 pathologists. Fifty-eight pathologists completed the questionnaire (55% response rate). The most important criteria for a consultant were knowledge and expertise. Accessibility, turnaround time, and teaching (providing explanation about the case) were selected next for choosing a consultant. The 2 factors that contributed to avoiding a consultant were expensive workup and changing the diagnosis. Open questions about "definition of best/worst consultant," "when to change the consultant," and "if the criteria for consultant have changed over time" provided additional valuable information.
CONCLUSIONS
Accessibility, short turnaround time, and teaching are the most important reasons for selecting a consultant. Performing an expensive workup and being in the habit of changing the diagnosis are the factors that make a consultant less favorable.
Topics: Consultants; Humans; Pathologists; Pathology, Surgical; Referral and Consultation; Surveys and Questionnaires
PubMed: 34075389
DOI: 10.1093/ajcp/aqab068 -
Turk Patoloji Dergisi 2018Molecular pathological analysis has an expanding role in patient diagnosis and management. The performance of these techniques relies on excellent laboratory procedures.... (Review)
Review
Molecular pathological analysis has an expanding role in patient diagnosis and management. The performance of these techniques relies on excellent laboratory procedures. However, the crucial step is obtaining the best samples for molecular analysis. Archiving and selection of these are the responsibilities of all pathologists even if they are not working at a center with molecular pathological facilities. This review focuses on the features of different types of materials for molecular pathological analysis. Many steps that might affect the results, including communication between the pathologist and the oncology team, features of different types of materials (cytological, tissue blocks, biopsy, circulating tumor cells (CTCs) and cell-free circulating nucleic acids), effects of tissue processing, methods for selecting the best material, and tissue saving and tumor enrichment methods are discussed. The procedures for referral to a center for molecular pathological analysis are also mentioned. Awareness of the importance of the cytopathological and histopathological material of the patients for future molecular pathological analysis by pathologists is of the utmost importance.
Topics: Humans; Pathologists; Pathology, Clinical; Pathology, Molecular
PubMed: 29235614
DOI: 10.5146/tjpath.2017.01420 -
Pathologie (Heidelberg, Germany) Sep 2022Accreditation in pathology and neuropathology in Germany is based on the requirements of the DIN EN ISO/IEC 17020 standard, which defines general requirements for the... (Review)
Review
Accreditation in pathology and neuropathology in Germany is based on the requirements of the DIN EN ISO/IEC 17020 standard, which defines general requirements for the bodies performing inspections. In Germany, the national accreditation body of the Federal Republic of Germany (DAkkS) assumes the official task of accreditation. Evaluation in this area is based on a comprehensive assessment by board-certified pathologists (inspection) and is focused on correct diagnosis. The requirement of the DIN EN ISO 15189 standard on quality and competence in medical laboratories is taken into account. In Germany, more than 100 institutes of pathology and neuropathology have successfully performed the accreditation process. Interest in accreditation has been significantly growing in recent months, especially due to changes in the European law with the "Verordnung EO 2017/746 des Europäischen Parlaments und des Rates vom 05. April 2017 über Invitro-Diagnostica".However, accreditation is a demanding process that requires very good preparation. Necessary resources for accreditation should be secured before starting the process. Important framework conditions and requirements, but also some challenges that can arise during the accreditation process, are summarized in this article.
Topics: Academies and Institutes; Accreditation; Certification; Germany; Neuropathology
PubMed: 35925314
DOI: 10.1007/s00292-022-01097-x