1
|
Levy JJ, Chan N, Marotti JD, Kerr DA, Gutmann EJ, Glass RE, Dodge CP, Suriawinata AA, Christensen BC, Liu X, Vaickus LJ. Large-scale validation study of an improved semiautonomous urine cytology assessment tool: AutoParis-X. Cancer Cytopathol 2023; 131:637-654. [PMID: 37377320 DOI: 10.1002/cncy.22732] [Citation(s) in RCA: 3] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/01/2023] [Revised: 05/11/2023] [Accepted: 05/12/2023] [Indexed: 06/29/2023]
Abstract
BACKGROUND Adopting a computational approach for the assessment of urine cytology specimens has the potential to improve the efficiency, accuracy, and reliability of bladder cancer screening, which has heretofore relied on semisubjective manual assessment methods. As rigorous, quantitative criteria and guidelines have been introduced for improving screening practices (e.g., The Paris System for Reporting Urinary Cytology), algorithms to emulate semiautonomous diagnostic decision-making have lagged behind, in part because of the complex and nuanced nature of urine cytology reporting. METHODS In this study, the authors report on the development and large-scale validation of a deep-learning tool, AutoParis-X, which can facilitate rapid, semiautonomous examination of urine cytology specimens. RESULTS The results of this large-scale, retrospective validation study indicate that AutoParis-X can accurately determine urothelial cell atypia and aggregate a wide variety of cell-related and cluster-related information across a slide to yield an atypia burden score, which correlates closely with overall specimen atypia and is predictive of Paris system diagnostic categories. Importantly, this approach accounts for challenges associated with the assessment of overlapping cell cluster borders, which improve the ability to predict specimen atypia and accurately estimate the nuclear-to-cytoplasm ratio for cells in these clusters. CONCLUSIONS The authors developed a publicly available, open-source, interactive web application that features a simple, easy-to-use display for examining urine cytology whole-slide images and determining the level of atypia in specific cells, flagging the most abnormal cells for pathologist review. The accuracy of AutoParis-X (and other semiautomated digital pathology systems) indicates that these technologies are approaching clinical readiness and necessitates full evaluation of these algorithms in head-to-head clinical trials.
Collapse
Affiliation(s)
- Joshua J Levy
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA
- Department of Dermatology, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA
- Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
- Program in Quantitative Biomedical Sciences, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Natt Chan
- Program in Quantitative Biomedical Sciences, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Jonathan D Marotti
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA
- Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Darcy A Kerr
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA
- Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Edward J Gutmann
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA
- Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Ryan E Glass
- Department of Pathology, University of Pittsburgh Medical Center East, Pittsburgh, Pennsylvania, USA
| | | | - Arief A Suriawinata
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA
- Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Brock C Christensen
- Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
- Department of Molecular and Systems Biology, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
- Department of Community and Family Medicine, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Xiaoying Liu
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA
- Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Louis J Vaickus
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA
- Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| |
Collapse
|
2
|
Deep learning for computational cytology: A survey. Med Image Anal 2023; 84:102691. [PMID: 36455333 DOI: 10.1016/j.media.2022.102691] [Citation(s) in RCA: 7] [Impact Index Per Article: 7.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/08/2022] [Revised: 10/22/2022] [Accepted: 11/09/2022] [Indexed: 11/16/2022]
Abstract
Computational cytology is a critical, rapid-developing, yet challenging topic in medical image computing concerned with analyzing digitized cytology images by computer-aided technologies for cancer screening. Recently, an increasing number of deep learning (DL) approaches have made significant achievements in medical image analysis, leading to boosting publications of cytological studies. In this article, we survey more than 120 publications of DL-based cytology image analysis to investigate the advanced methods and comprehensive applications. We first introduce various deep learning schemes, including fully supervised, weakly supervised, unsupervised, and transfer learning. Then, we systematically summarize public datasets, evaluation metrics, versatile cytology image analysis applications including cell classification, slide-level cancer screening, nuclei or cell detection and segmentation. Finally, we discuss current challenges and potential research directions of computational cytology.
Collapse
|
3
|
Meijering E. A bird's-eye view of deep learning in bioimage analysis. Comput Struct Biotechnol J 2020; 18:2312-2325. [PMID: 32994890 PMCID: PMC7494605 DOI: 10.1016/j.csbj.2020.08.003] [Citation(s) in RCA: 58] [Impact Index Per Article: 14.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/28/2020] [Revised: 07/26/2020] [Accepted: 08/01/2020] [Indexed: 02/07/2023] Open
Abstract
Deep learning of artificial neural networks has become the de facto standard approach to solving data analysis problems in virtually all fields of science and engineering. Also in biology and medicine, deep learning technologies are fundamentally transforming how we acquire, process, analyze, and interpret data, with potentially far-reaching consequences for healthcare. In this mini-review, we take a bird's-eye view at the past, present, and future developments of deep learning, starting from science at large, to biomedical imaging, and bioimage analysis in particular.
Collapse
Affiliation(s)
- Erik Meijering
- School of Computer Science and Engineering & Graduate School of Biomedical Engineering, University of New South Wales, Sydney, Australia
| |
Collapse
|
4
|
PathoSpotter-K: A computational tool for the automatic identification of glomerular lesions in histological images of kidneys. Sci Rep 2017; 7:46769. [PMID: 28436482 PMCID: PMC5402276 DOI: 10.1038/srep46769] [Citation(s) in RCA: 29] [Impact Index Per Article: 4.1] [Reference Citation Analysis] [Abstract] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/27/2017] [Accepted: 03/27/2017] [Indexed: 11/20/2022] Open
Abstract
PathoSpotter is a computational system designed to assist pathologists in teaching about and researching kidney diseases. PathoSpotter-K is the version that was developed to detect nephrological lesions in digital images of kidneys. Here, we present the results obtained using the first version of PathoSpotter-K, which uses classical image processing and pattern recognition methods to detect proliferative glomerular lesions with an accuracy of 88.3 ± 3.6%. Such performance is only achieved by similar systems if they use images of cell in contexts that are much less complex than the glomerular structure. The results indicate that the approach can be applied to the development of systems designed to train pathology students and to assist pathologists in determining large-scale clinicopathological correlations in morphological research.
Collapse
|
5
|
Kirleis MA, Mathews SA, Verbarg J, Erickson JS, Piqué A. Reconfigurable acquisition system with integrated optics for a portable flow cytometer. THE REVIEW OF SCIENTIFIC INSTRUMENTS 2013; 84:115109. [PMID: 24289439 DOI: 10.1063/1.4831835] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/02/2023]
Abstract
Portable and inexpensive scientific instruments that are capable of performing point of care diagnostics are needed for applications such as disease detection and diagnosis in resource-poor settings, for water quality and food supply monitoring, and for biosurveillance activities in autonomous vehicles. In this paper, we describe the development of a compact flow cytometer built from three separate, customizable, and interchangeable modules. The instrument as configured in this work is being developed specifically for the detection of selected Centers for Disease Control (CDC) category B biothreat agents through a bead-based assay: E. coli O157:H7, Salmonella, Listeria, and Shigella. It has two-color excitation, three-color fluorescence and light scattering detection, embedded electronics, and capillary based flow. However, these attributes can be easily modified for other applications such as cluster of differentiation 4 (CD4) counting. Proof of concept is demonstrated through a 6-plex bead assay with the results compared to a commercially available benchtop-sized instrument.
Collapse
Affiliation(s)
- Matthew A Kirleis
- Naval Research Laboratory, 4555 Overlook Ave. SW, Washington, DC 20375, USA
| | | | | | | | | |
Collapse
|