1
|
Liu Y, Bilodeau E, Pollack B, Batmanghelich K. Automated detection of premalignant oral lesions on whole slide images using convolutional neural networks. Oral Oncol 2022; 134:106109. [PMID: 36126604 DOI: 10.1016/j.oraloncology.2022.106109] [Citation(s) in RCA: 6] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/04/2022] [Revised: 08/12/2022] [Accepted: 08/29/2022] [Indexed: 11/30/2022]
Abstract
INTRODUCTION Oral epithelial dysplasia (OED) is a precursor lesion to oral squamous cell carcinoma, a disease with a reported overall survival rate of 56 percent across all stages. Accurate detection of OED is critical as progression to oral cancer can be impeded with complete excision of premalignant lesions. However, previous research has demonstrated that the task of grading of OED, even when performed by highly trained experts, is subject to high rates of reader variability and misdiagnosis. Thus, our study aims to develop a convolutional neural network (CNN) model that can identify regions suspicious for OED whole-slide pathology images. METHODS During model development, we optimized key training hyperparameters including loss function on 112 pathologist annotated cases between the training and validation sets. Then, we compared OED segmentation and classification metrics between two well-established CNN architectures for medical imaging, DeepLabv3+ and UNet++. To further assess generalizability, we assessed case-level performance of a held-out test set of 44 whole-slide images. RESULTS DeepLabv3+ outperformed UNet++ in overall accuracy, precision, and segmentation metrics in a 4-fold cross validation study. When applied to the held-out test set, our best performing DeepLabv3+ model achieved an overall accuracy and F1-Score of 93.3 percent and 90.9 percent, respectively. CONCLUSION The present study trained and implemented a CNN-based deep learning model for identification and segmentation of oral epithelial dysplasia (OED) with reasonable success. Computer assisted detection was shown to be feasible in detecting premalignant/precancerous oral lesions, laying groundwork for eventual clinical implementation.
Collapse
Affiliation(s)
- Yingci Liu
- University of Pittsburgh, Department of Biomedical Informatics, 5607, Baum Boulevard, Pittsburgh, PA 15206, USA; Rutgers School of Dental Medicine, 110, Bergen St, Newark, NJ 07101, USA.
| | - Elizabeth Bilodeau
- University of Pittsburgh School of Dental Medicine, 3501 Terrace St., Pittsburgh, PA 15206, USA
| | - Brian Pollack
- University of Pittsburgh, Department of Biomedical Informatics, 5607, Baum Boulevard, Pittsburgh, PA 15206, USA
| | - Kayhan Batmanghelich
- University of Pittsburgh, Department of Biomedical Informatics, 5607, Baum Boulevard, Pittsburgh, PA 15206, USA
| |
Collapse
|
2
|
Odell E, Kujan O, Warnakulasuriya S, Sloan P. Oral epithelial dysplasia: Recognition, grading and clinical significance. Oral Dis 2021; 27:1947-1976. [PMID: 34418233 DOI: 10.1111/odi.13993] [Citation(s) in RCA: 93] [Impact Index Per Article: 31.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/07/2021] [Revised: 07/14/2021] [Accepted: 07/31/2021] [Indexed: 12/29/2022]
Abstract
Histopathological grading of epithelial dysplasia remains the principal laboratory method for assessing the risk of malignant transformation in oral potentially malignant disorders (OPMDs). Current views on the molecular pathogenesis and histological interpretation of the features of epithelial dysplasia are described, and the use of grading systems for epithelial dysplasia is discussed. Changes to the current 2017 WHO criteria for diagnosis are proposed with emphasis on the architectural features of epithelial dysplasia. The predictive values of three-grade and binary systems are summarised, and categories of epithelial dysplasia are reviewed, including lichenoid and verrucous lesions, keratosis of unknown significance, HPV-associated dysplasia, differentiated and basaloid epithelial dysplasia. The implications of finding epithelial dysplasia in an oral biopsy for clinical management are discussed from the pathologists' viewpoint.
Collapse
Affiliation(s)
- Edward Odell
- King's College London and Head and Neck Pathology Guy's Hospital, London, UK
| | - Omar Kujan
- UWA Dental School, The University of Western Australia, Perth, WA, Australia
| | - Saman Warnakulasuriya
- Faculty of Dentistry, Oral and Craniofacial Sciences King's College London and The WHO Collaborating Centre for Oral Cancer, King's College London, London, UK
| | - Philip Sloan
- School of Dental Sciences, Faculty of Medical Sciences, Newcastle University, Newcastle upon Tyne, UK.,Department of Cellular Pathology, Newcastle upon Tyne Hospitals NHS Foundation Trust, Newcastle upon Tyne, UK.,Chief Histopathologist, AMLo Biosciences, Newcastle upon Tyne, UK
| |
Collapse
|
3
|
Pritzker KPH, Darling MR, Hwang JTK, Mock D. Oral Potentially Malignant Disorders (OPMD): What is the clinical utility of dysplasia grade? Expert Rev Mol Diagn 2021; 21:289-298. [PMID: 33682567 DOI: 10.1080/14737159.2021.1898949] [Citation(s) in RCA: 9] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/07/2023]
Abstract
INTRODUCTION Oral epithelial dysplasia is considered a potential histologic precursor of subsequent squamous cell cancer. As standard clinical practice, pathologists grade dysplasia to assess risk for progression to malignancy. Except for the most advanced grade, severe dysplasia, dysplasia grading has failed to correlate well with the risk to develop invasive cancer. The questions of what process dysplasia grading best represents and what clinical utility dysplasia grading may have are explored. AREAS COVERED This narrative review is based on PubMed search with emphasis on papers since 2010. Epithelial dysplasia as a precursor lesion of cancer and dysplasia grading as a risk assessment tool for progression to cancer are discussed. The close clinical association of dysplasia with known carcinogens, alcohol, and tobacco products is presented. EXPERT OPINION Oral epithelial dysplasia is often, associated with prolonged exposure to tobacco and alcohol products. With reduction of carcinogen exposure, dysplasia is known to regress in some cases. It is proposed that histologic dysplasia grade together with macroscopic images of dysplastic clinical lesions be used as an educational tool to incentivize patients to reduce their known carcinogen exposure. This strategy has the potential to reduce lesion progression thereby reducing the disease burden of oral cancer.
Collapse
Affiliation(s)
- Kenneth P H Pritzker
- Professor Emeritus, Laboratory Medicine and Pathobiology; Surgery University of Toronto, Toronto, Ontario, Canada.,Proteocyte Diagnostics Inc., Toronto, Canada.,Department of Pathology and Laboratory Medicine, Pathology & Laboratory Medicine Mount Sinai Hospital, Toronto, Canada
| | - Mark R Darling
- Professor, Department of Pathology and Laboratory Medicine, Schulich Faculty of Medicine and Dentistry, Western University London Ontario, Canada
| | | | - David Mock
- Department of Pathology and Laboratory Medicine, Pathology & Laboratory Medicine Mount Sinai Hospital, Toronto, Canada.,Professor, Pathology/Oral Medicine & Dean Emeritus, Faculty of Dentistry, University of Toronto, Toronto, Ontario, Canada.,Department of Dentistry, Dentistry Mount Sinai Hospital, Toronto, Canada
| |
Collapse
|
4
|
Santoshi CK, Kumar JV, Bhagirath PV, Vinay BH, Prakash YJ. Morphometric analysis of basal cells of oral epithelium in predicting malignant transformation of oral potentially malignant disorders in patients with tobacco chewing habit. J Oral Maxillofac Pathol 2020; 24:579. [PMID: 33967506 PMCID: PMC8083430 DOI: 10.4103/jomfp.jomfp_55_20] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/07/2020] [Accepted: 10/07/2020] [Indexed: 11/04/2022] Open
Abstract
BACKGROUND Oral potentially malignant disorders (OPMDs) are a heterogenous group of disorders which precede the development of oral cancer. These are characterized by increased risk of malignant transformation to Oral cancer. AIMS & OBJECTIVES In this study, an attempt has been made to assess the morphological alterations of the nuclei of the basal cells in OPMDs and oral squamous cell carcinoma (OSCC). Our objective was to compare the alterations and to assess the predictive factor of such alterations of basal cells in malignant transformation of OPMDs to OSCC. MATERIALS AND METHODS This was a retrospective study conducted on tissue sections of 150 formalin fixed, paraffin embedded blocks obtained from the archives. The specimens were grouped into OSCC group (n= 50) and OPMDs (n= 100). Nuclear features were evaluated using computer- assisted microscopic image analysis. One- way ANOVA analysis was done to verify the difference between the groups for all variables. RESULTS Our results showed statistically significant difference for all parameters between the groups. Among OPMDs, leukoplakia showed significant increase in nuclear area, nuclear perimeter, Nuclear/Cytoplasm (N/C) ratio, density and loss of polarity, while OSF showed significance with only perimeter, density and loss of polarity. CONCLUSION Based on findings of present study, it is concluded that measurements using computer- aided morphometric analysis may provide an objective means for predicting the malignant transformation OPMDs to OSCC. Among OPMDs, Leukoplakia has a higher chance of malignant transformation than OSF.
Collapse
Affiliation(s)
- Chatraspalli Krishna Santoshi
- Department of Oral and Maxillofacial Pathology, Panineeya Mahavidyalaya Institute of Dental Sciences and Research Centre, Hyderabad, Telangana, India
| | - Jogishetty Vijay Kumar
- Department of Oral and Maxillofacial Pathology, Panineeya Mahavidyalaya Institute of Dental Sciences and Research Centre, Hyderabad, Telangana, India
| | - Pacha Venkat Bhagirath
- Department of Oral and Maxillofacial Pathology, Panineeya Mahavidyalaya Institute of Dental Sciences and Research Centre, Hyderabad, Telangana, India
| | - B Hari Vinay
- Department of Oral and Maxillofacial Pathology, Panineeya Mahavidyalaya Institute of Dental Sciences and Research Centre, Hyderabad, Telangana, India
| | - Y John Prakash
- Department of Oral and Maxillofacial Pathology, Panineeya Mahavidyalaya Institute of Dental Sciences and Research Centre, Hyderabad, Telangana, India
| |
Collapse
|
5
|
Gonzaga AKG, Mafra RP, da Silva LP, de Almeida Freitas R, de Souza LB, Pinto LP. Actinic cheilitis: Morphometric parameters and its relationship with the degree of epithelial dysplasia. Acta Histochem 2020; 122:151452. [PMID: 31558280 DOI: 10.1016/j.acthis.2019.151452] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/06/2019] [Revised: 09/13/2019] [Accepted: 09/16/2019] [Indexed: 12/31/2022]
Abstract
Actinic cheilitis (AC) is a potentially malignant lesion caused by chronic sun exposure. This study aimed to evaluate the relationship between the degree of epithelial dysplasia and morphometric findings in AC. Sixty-eight slides of AC cases were selected and classified according to the grade of epithelial dysplasia, following morphologic criteria of World Health Organization. For morphometric analysis, the slides were scanned and images were analyzed using Pannoramic Viewer software. We obtained vertical measurements of the parameters: thicknesses of the keratin layer, lamina propria and zone of solar elastosis in three selected fields. Thirty-seven (54.4%) of the analyzed cases were classified as none/mild dysplasia and 31 (45.6%) as moderate/severe epithelial dysplasia. Cases with a moderate/severe dysplasia exhibited a thicker layer of keratin (median = 0.055 mm) than none/mild dysplasia (median = 0.045 mm) (p = 0.033). No significant differences in the thicknesses of lamina propria and zone of solar elastosis were observed according to the grade of epithelial dysplasia. A positive significant correlation between keratin layer and lamina propria thicknesses was found (p = 0.019). Based on our findings, rigorous clinical follow-up should be recommended for patients whose histopathological examination shows a greater thickness of the keratin layer.
Collapse
|
6
|
Nag R, Kumar Das R. Analysis of images for detection of oral epithelial dysplasia: A review. Oral Oncol 2018; 78:8-15. [PMID: 29496062 DOI: 10.1016/j.oraloncology.2018.01.003] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/09/2017] [Revised: 01/02/2018] [Accepted: 01/07/2018] [Indexed: 10/18/2022]
Abstract
This paper provides a review of various image analysis approaches that have been previously used for recognition of dysplasia in images of the epithelium of the oral cavity. This domain has become especially admissible with the uncovering of the importance of image analysis which can probably be an aid to subjective diagnosis by histopathologists. Oral malignancy is a rampant form of cancer found among people of the Indian subcontinent due to various deleterious habits like consumption of tobacco, areca nut, betel leaf etc. Oral Submucous Fibrosis, a precancer, whose pathological category falls between normal epithelium and epithelial dysplasia, is caused because of these habits and can ultimately lead to oral cancer. Hence early detection of this condition is necessary. Image analysis methods for this purpose have an enormous potential which can also reduce the heavy workload of pathologists and to refine the criterion of interpretation. This paper starts with a critique of statistics of oral carcinoma in India and distribution of cancer in intra-oral sites and moves on to its causes and diagnostic approaches including causative agents, problems in curative approach and importance of image analysis in cancer detection. The various image analysis methods to appraise the cytological and architectural changes accompanied by Oral Epithelial Dysplasia in the images of the oral epithelial region have been described in relation to 2005 WHO Classification System and it was found that in future, analysis of images based on the mentioned methods has the potential in better interpretation and diagnosis of oral carcinoma.
Collapse
Affiliation(s)
- Reetoja Nag
- Centre for Biomaterials, Cellular, and Molecular Theranostics, VIT University, Vellore 632014, India.
| | - Raunak Kumar Das
- Centre for Biomaterials, Cellular, and Molecular Theranostics, VIT University, Vellore 632014, India; School of Biotechnology and Biomolecular Sciences, University of New South Wales, Sydney, Australia.
| |
Collapse
|