-
The American Journal of Pathology Sep 2019With the rapid development of image scanning techniques and visualization software, whole slide imaging (WSI) is becoming a routine diagnostic method. Accelerating... (Review)
Review
With the rapid development of image scanning techniques and visualization software, whole slide imaging (WSI) is becoming a routine diagnostic method. Accelerating clinical diagnosis from pathology images and automating image analysis efficiently and accurately remain significant challenges. Recently, deep learning algorithms have shown great promise in pathology image analysis, such as in tumor region identification, metastasis detection, and patient prognosis. Many machine learning algorithms, including convolutional neural networks, have been proposed to automatically segment pathology images. Among these algorithms, segmentation deep learning algorithms such as fully convolutional networks stand out for their accuracy, computational efficiency, and generalizability. Thus, deep learning-based pathology image segmentation has become an important tool in WSI analysis. In this review, the pathology image segmentation process using deep learning algorithms is described in detail. The goals are to provide quick guidance for implementing deep learning into pathology image analysis and to provide some potential ways of further improving segmentation performance. Although there have been previous reviews on using machine learning methods in digital pathology image analysis, this is the first in-depth review of the applications of deep learning algorithms for segmentation in WSI analysis.
Topics: Algorithms; Deep Learning; Diagnostic Imaging; Humans; Image Processing, Computer-Assisted; Pathology, Clinical
PubMed: 31199919
DOI: 10.1016/j.ajpath.2019.05.007 -
JAMA Dec 2017Application of deep learning algorithms to whole-slide pathology images can potentially improve diagnostic accuracy and efficiency. (Comparative Study)
Comparative Study
IMPORTANCE
Application of deep learning algorithms to whole-slide pathology images can potentially improve diagnostic accuracy and efficiency.
OBJECTIVE
Assess the performance of automated deep learning algorithms at detecting metastases in hematoxylin and eosin-stained tissue sections of lymph nodes of women with breast cancer and compare it with pathologists' diagnoses in a diagnostic setting.
DESIGN, SETTING, AND PARTICIPANTS
Researcher challenge competition (CAMELYON16) to develop automated solutions for detecting lymph node metastases (November 2015-November 2016). A training data set of whole-slide images from 2 centers in the Netherlands with (n = 110) and without (n = 160) nodal metastases verified by immunohistochemical staining were provided to challenge participants to build algorithms. Algorithm performance was evaluated in an independent test set of 129 whole-slide images (49 with and 80 without metastases). The same test set of corresponding glass slides was also evaluated by a panel of 11 pathologists with time constraint (WTC) from the Netherlands to ascertain likelihood of nodal metastases for each slide in a flexible 2-hour session, simulating routine pathology workflow, and by 1 pathologist without time constraint (WOTC).
EXPOSURES
Deep learning algorithms submitted as part of a challenge competition or pathologist interpretation.
MAIN OUTCOMES AND MEASURES
The presence of specific metastatic foci and the absence vs presence of lymph node metastasis in a slide or image using receiver operating characteristic curve analysis. The 11 pathologists participating in the simulation exercise rated their diagnostic confidence as definitely normal, probably normal, equivocal, probably tumor, or definitely tumor.
RESULTS
The area under the receiver operating characteristic curve (AUC) for the algorithms ranged from 0.556 to 0.994. The top-performing algorithm achieved a lesion-level, true-positive fraction comparable with that of the pathologist WOTC (72.4% [95% CI, 64.3%-80.4%]) at a mean of 0.0125 false-positives per normal whole-slide image. For the whole-slide image classification task, the best algorithm (AUC, 0.994 [95% CI, 0.983-0.999]) performed significantly better than the pathologists WTC in a diagnostic simulation (mean AUC, 0.810 [range, 0.738-0.884]; P < .001). The top 5 algorithms had a mean AUC that was comparable with the pathologist interpreting the slides in the absence of time constraints (mean AUC, 0.960 [range, 0.923-0.994] for the top 5 algorithms vs 0.966 [95% CI, 0.927-0.998] for the pathologist WOTC).
CONCLUSIONS AND RELEVANCE
In the setting of a challenge competition, some deep learning algorithms achieved better diagnostic performance than a panel of 11 pathologists participating in a simulation exercise designed to mimic routine pathology workflow; algorithm performance was comparable with an expert pathologist interpreting whole-slide images without time constraints. Whether this approach has clinical utility will require evaluation in a clinical setting.
Topics: Algorithms; Breast Neoplasms; Female; Humans; Lymphatic Metastasis; Machine Learning; Pathologists; Pathology, Clinical; ROC Curve
PubMed: 29234806
DOI: 10.1001/jama.2017.14585 -
Zhonghua Bing Li Xue Za Zhi = Chinese... Jan 2022With the technological progresses and applications of human genome sequencing, bioinformatics analysis and data mining, and molecular pathology and artificial...
With the technological progresses and applications of human genome sequencing, bioinformatics analysis and data mining, and molecular pathology and artificial intelligence-assisted pathological diagnosis, the development of clinical medicine is moving towards the era of precision diagnosis and treatment. In the context of this era, the traditional diagnostic pathology is facing unprecedented opportunities and challenges in our history and is striving towards the "next-generation diagnostic pathology" (NGDP). NGDP is based on histomorphology and clinical data, and characterized by the combination of molecular detection and bioinformatics analysis, intelligent sampling and process quality control, intelligent diagnosis and remote consultation, lesion visualization and "non-invasive" pathology as well as other innovative cutting edge interdisciplinary technologies. The NGDP reports will include the results from multi-omics and cross-scale integrated diagnosis for final diagnosis. NGDP will also be applied for predicting disease progression and outcomes, and determining optional therapeutics as well as assessing treatment responses, so that a novel "golden standard" of disease diagnosis can be established. In the near fature, it is necessary to stimulate the innovative vitality of pathology disciplines, accelerate the maturity and application for NGDP, update the theory and technical system of pathology, and perform its important applicable role in the prevention, diagnosis, treatment of diseases so that the futher development of clinical medicine will be promoted and the strategy for maintenance of being healthy in China will be served.
Topics: Artificial Intelligence; China; Computational Biology; Humans; Pathology, Molecular
PubMed: 34979745
DOI: 10.3760/cma.j.cn112151-20211005-00726 -
Diagnostic Cytopathology Jan 2023In the era of personalized medicine, molecular testing plays a critical role in patient care. The rapid advance of molecular techniques, especially next-generation...
In the era of personalized medicine, molecular testing plays a critical role in patient care. The rapid advance of molecular techniques, especially next-generation sequencing, makes molecular diagnosis feasible in daily practice. Molecular testing can be used as a valuable ancillary test to increase diagnostic sensitivity and specificity, especially in small biopsy or cytology samples. In addition, molecular testing plays an important role in selecting patients for appropriate treatment by detecting therapeutic and predictive biomarkers in tissue or cytology samples. Molecular studies can be applied in all cytology samples, sometimes with better results than histology. As molecular testing has become essential for patient care and is often requested to be performed in cytology samples, it is critical for cytopathologists to understand the basics of molecular diagnostic methods, indications for molecular testing, and how to best utilize different cytologic samples for this purpose. In this special issue, experts in various areas of cytopathology and molecular pathology review the literature and discuss the basics of molecular techniques and the application of molecular testing in various types of cytology samples. It is our hope that after reading the articles in this special issue, the readers can know better about the possibilities of molecular cytology, a very exciting field of pathology.
Topics: Humans; Pathology, Molecular; Molecular Diagnostic Techniques
PubMed: 36367273
DOI: 10.1002/dc.25071 -
Pathology Jan 2019This review is an evidence-based summary of digital pathology: past, present and future. It discusses digital surgical pathology and the cytopathology digitisation... (Review)
Review
This review is an evidence-based summary of digital pathology: past, present and future. It discusses digital surgical pathology and the cytopathology digitisation challenge as well as the performance of digital histopathology and cytopathology as a diagnostic tool, particularly in contrast to user perceptions. Time and cost efficiency of digital pathology, learning curves, education and quality assurance, with the importance of validation of systems, is emphasised. The review concludes with a discussion of digital pathology as a source of 'big data' and where this might lead pathologists in the digital pathology future.
Topics: Cytodiagnosis; Humans; Microscopy; Pathology, Surgical
PubMed: 30522785
DOI: 10.1016/j.pathol.2018.10.011 -
The Veterinary Clinics of North... Apr 2020The assessment of blood analytes in racehorses can provide useful data on performance and health. The horses' adaptive responses to training that occur to optimize... (Review)
Review
The assessment of blood analytes in racehorses can provide useful data on performance and health. The horses' adaptive responses to training that occur to optimize performance should be considered when interpreting alterations seen on laboratory results. Similarly, the alterations observed in laboratory test results can identify subclinical and clinical disease and be helpful for identifying organ dysfunction and, in many cases, monitoring progress and response to treatment. This article discusses hematologic and biochemical tests that are important in the evaluation of performance and health in racehorses.
Topics: Animals; Horse Diseases; Horses; Pathology, Clinical; Physical Conditioning, Animal
PubMed: 31992502
DOI: 10.1016/j.cveq.2019.12.004 -
American Journal of Clinical Pathology May 2022To demonstrate how the educational presentation and targeted review of cases with discrepant interpretive findings between pathologists can raise awareness for specific... (Review)
Review
OBJECTIVES
To demonstrate how the educational presentation and targeted review of cases with discrepant interpretive findings between pathologists can raise awareness for specific diagnostic errors through identification of common overarching patterns of error in interpretive pathology.
METHODS
We performed a review of 147 surgical pathology and cytopathology cases of discordances from 23 PowerPoint presentations presented between 2010 and 2017. Pathologists and pathology residents, blinded from the official interpretations, were presented each case and surveyed for their own diagnostic assessments. Survey results were compared with the final/correct interpretations of the signing pathologists.
RESULTS
Of the 134 cases with available survey results, there were 87 (64.9%) for which most survey respondents proposed a diagnostic interpretation concordant with the final/correct diagnosis. There were 37 (27.6%) cases for which most survey responses were either wholly or partially discordant with the final/correct diagnosis. For 10 (7.5%) cases, there were equal numbers of concordant and discordant survey responses.
CONCLUSIONS
Our analyses of the cases with frequent erroneous diagnoses reveal common patterns of error that are widely applicable and outline specific error-prone interpretive tendencies. Greater awareness for these tendencies, highlighted by presentation of discordant cases, can improve the quality of diagnostic pathology services.
Topics: Diagnostic Errors; Humans; Pathology, Surgical
PubMed: 35512255
DOI: 10.1093/ajcp/aqab190 -
Wiener Medizinische Wochenschrift (1946) Sep 2021Dementia is the clinical consequence of various neurological disorders with a multitude of etiologies. Precise knowledge of the underlying pathologies is essential for... (Review)
Review
Dementia is the clinical consequence of various neurological disorders with a multitude of etiologies. Precise knowledge of the underlying pathologies is essential for an accurate treatment of patients and for the development of suitable disease biomarkers. A definite diagnosis of many of the disorders, particularly for neurodegenerative ones, can only be made after a thorough postmortem neuropathological examination. This highlights the importance of performing a brain autopsy and the relevance of a close interaction between clinicians, neuroimaging disciplines and neuropathologists as well as with basic researchers. This article aims to give a brief overview on the neuropathology of dementia focusing on neurodegenerative diseases, to further facilitate interdisciplinary collaboration.
Topics: Autopsy; Brain; Dementia; Humans; Neurodegenerative Diseases; Neuropathology
PubMed: 34129141
DOI: 10.1007/s10354-021-00848-4 -
Histopathology Jan 2024Currently, lung cancer is treated by the highest number of therapeutic options and the benefits are based on multiple large-scale sequencing studies, translational... (Review)
Review
Currently, lung cancer is treated by the highest number of therapeutic options and the benefits are based on multiple large-scale sequencing studies, translational research and new drug development, which has promoted our understanding of the molecular pathology of lung cancer. According to the driver alterations, different characteristics have been revealed, such as differences in ethnic prevalence, median age and alteration patterns. Consequently, beyond traditional chemoradiotherapy, molecular-targeted therapy and treatment with immune check-point inhibitors (ICI) also became available major therapeutic options. Interestingly, clinical results suggest that the recently established therapies target distinct lung cancer proportions, particularly between the EGFR/ALK and PD-1/PD-L1-positive subsets, e.g. the kinase inhibitors target driver mutation-positive tumours, whereas driver mutation-negative tumours respond to ICI treatment. These therapeutic efficacy-related differences might be explained by the molecular pathogenesis of lung cancer. Addictive driver mutations promote tumour formation with powerful transformation performance, resulting in a low tumour mutation burden, reduced immune surveillance, and subsequent poor response to ICIs. In contrast, regular tobacco smoke exposure repeatedly injures the proximal airway epithelium, leading to accumulated genetic alterations. In the latter pathway, overgrowth due to alteration and immunological exclusion against neoantigens is initially balanced. However, tumours could be generated from certain clones that outcompete immunological exclusion and outgrow the others. Consequently, this cancer type responds to immune check-point treatment. These pathogenic differences are explained well by the two-compartment model, focusing upon the anatomical and functional composition of distinct cellular components between the terminal respiratory unit and the air-conducting system.
Topics: Humans; Carcinoma, Non-Small-Cell Lung; Pathology, Molecular; Lung Neoplasms; Carcinoma; Mutation; B7-H1 Antigen
PubMed: 37936491
DOI: 10.1111/his.15080 -
Seminars in Cancer Biology Jul 2021Deep Learning (DL) algorithms are a set of techniques that exploit large and/or complex real-world datasets for cross-domain and cross-discipline prediction and... (Review)
Review
Deep Learning (DL) algorithms are a set of techniques that exploit large and/or complex real-world datasets for cross-domain and cross-discipline prediction and classification tasks. DL architectures excel in computer vision tasks, and in particular image processing and interpretation. This has prompted a wave of disruptingly innovative applications in medical imaging, where DL strategies have the potential to vastly outperform human experts. This is particularly relevant in the context of histopathology, where whole slide imaging (WSI) of stained tissue in conjuction with DL algorithms for their interpretation, selection and cancer staging are beginning to play an ever increasing role in supporting human operators in visual assessments. This has the potential to reduce everyday workload as well as to increase precision and reproducibility across observers, centers, staining techniques and even pathologies. In this paper we introduce the most common DL architectures used in image analysis, with a focus on histopathological image analysis in general and in breast histology in particular. We briefly review how, state-of-art DL architectures compare to human performance on across a number of critical tasks such as mitotic count, tubules analysis and nuclear pleomorphism analysis. Also, the development of DL algorithms specialized to pathology images have been enormously fueled by a number of world-wide challenges based on large, multicentric image databases which are now publicly available. In turn, this has allowed most recent efforts to shift more and more towards semi-supervised learning methods, which provide greater flexibility and applicability. We also review all major repositories of manually labelled pathology images in breast cancer and provide an in-depth discussion of the challenges specific to training DL architectures to interpret WSI data, as well as a review of the state-of-the-art methods for interpretation of images generated from immunohistochemical analysis of breast lesions. We finally discuss the future challenges and opportunities which the adoption of DL paradigms is most likely to pose in the field of pathology for breast cancer detection, diagnosis, staging and prognosis. This review is intended as a comprehensive stepping stone into the field of modern computational pathology for a transdisciplinary readership across technical and medical disciplines.
Topics: Breast Neoplasms; Computational Biology; Deep Learning; Diagnostic Imaging; Female; Humans; Image Processing, Computer-Assisted; Pathology, Clinical
PubMed: 32818626
DOI: 10.1016/j.semcancer.2020.08.006