Oral potentially malignant disorders: Proceedings from an expert symposium

Warnakulasuriya, Saman; Lodi, Giovanni

doi:10.1111/odi.13999

Cited by 9 publications

(12 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Our dataset included every single region in the oral cavity that might have OPMDs. 40–42 Since the floor of the mouth and tongue are the most common areas for OC, 40 their study improved our knowledge and contributed to the literature significantly, however until all oral mucosal sites are involved, a proper CNN model cannot be used yet. Moreover, the significance of our approach is supported by clinically relevant development of multifocal OPMD, such as in the case of Proliferative verrucous leukoplakia which possess high risk of transforming into OC.…”

Section: Discussionmentioning

confidence: 99%

Deep convolutional neural network algorithm for the automatic segmentation of oral potentially malignant disorders and oral cancers

Ünsal

Chaurasia

Akkaya

et al. 2023

Proc Inst Mech Eng H

View full text Add to dashboard Cite

This study aimed to develop an algorithm to automatically segment the oral potentially malignant diseases (OPMDs) and oral cancers (OCs) of all oral subsites with various deep convolutional neural network applications. A total of 510 intraoral images of OPMDs and OCs were collected over 3 years (2006—2009). All images were confirmed both with patient records and histopathological reports. Following the labeling of the lesions the dataset was arbitrarily split, using random sampling in Python as the study dataset, validation dataset, and test dataset. Pixels were classified as the OPMDs and OCs with the OPMD/OC label and the rest as the background. U-Net architecture was used and the model with the best validation loss was chosen for the testing among the trained 500 epochs. Dice similarity coefficient (DSC) score was noted. The intra-observer ICC was found to be 0.994 while the inter-observer reliability was 0.989. The calculated DSC and validation accuracy across all clinical images were 0.697 and 0.805, respectively. Our algorithm did not maintain an excellent DSC due to multiple reasons for the detection of both OC and OPMDs in oral cavity sites. A better standardization for both 2D and 3D imaging (such as patient positioning) and a bigger dataset are required to improve the quality of such studies. This is the first study which aimed to segment OPMDs and OCs in all subsites of oral cavity which is crucial not only for the early diagnosis but also for higher survival rates.

show abstract

Section: Discussionmentioning

confidence: 99%

Deep convolutional neural network algorithm for the automatic segmentation of oral potentially malignant disorders and oral cancers

Ünsal

Chaurasia

Akkaya

et al. 2023

Proc Inst Mech Eng H

View full text Add to dashboard Cite

show abstract

“…This, in turn, correlates with the clinical observations that higher rates of MT were found in patients with atrophic OLP lesions ( 44 ). Overexpression of PCNA has also been reported to be associated with neoplastic progression in oral submucosa fibrosis, which is classified by WHO as an oral potentially malignant disorder ( 6 ). The observed correlation between increased PCNA expression and MT may be due to the fact that the higher the fraction of growth, the higher the chance to acquire mutations and to eventually develop malignant phenotypes.…”

Section: Discussionmentioning

confidence: 99%

“…The overall malignant transformation rate (MTR) of OLP varies between 0.5 -1.2 % ( 5 ) but has been reported to be as high as 2.8% ( 6 ). The MTR varies among different clinical types but generally higher in atrophic and/or erosive types.…”

Section: Introductionmentioning

confidence: 99%

Significance of immunohistochemistry biomarkers in prediction of malignant transformation of oral lichen planus: A systematic review

Al-Jamaei¹,

Subramanyam²,

Helder³

et al. 2022

Med Oral

View full text Add to dashboard Cite

Background Oral lichen planus (OLP) is a chronic inflammatory disorder with increased risk for malignant transformation. Biomarker validation is a pivotal step in moving newly discovered biomarkers towards clinical implementation. We performed a systematic review of studies on biomarkers related to OLP, wherein biomarkers have been described in at least two independent studies. Our aim was to determine whether any of these biomarkers might be promising in predicting the increased risk of malignant transformation of OLP. Material and Methods We searched the following databases until August 2021: PUBMED, EMBASE, and Web of Science. Due to high heterogeneity, a qualitative rather than quantitative assessment was conducted. Only proteins that consistently showed a significantly high level of expression in neoplastic tissues versus OLP in two or more publications were considered as promising markers. Results Initial database researches identified 1671, of which 24 articles were included in the final analysis. The most frequently reported proteins were p53, Bcl-2 and Ki-67, though there were controversies. PCNA and P21 were the only proteins that showed consistent evidence of clinical usefulness as cancer predictors to be considered as promising markers. Extensive methodological variations in the evaluation of expressions and statistical analyses of the included markers were observed, which hampered comparisons of the results. Conclusions Multiple levels of heterogeneity with a scarcity of high-quality studies were identified. PCNA and P21 were identified as promising predictive markers for evaluating the risk of malignant transformation of OLP, but they require further validation. The focus of future research on validation of predictive biomarkers of OLP should be considered as a high priority because it will accelerate the introduction of newly discovered markers into the clinical setting. Key words: Oral lichen planus, Immunohistochemistry, biomarkers, malignant transformation, PCNA, P21, p53, BCl-2.

show abstract

“…However, this is only a statistical effect, as the determination of DNA ploidy provides a more definitive result for a clinical measure than does doubtful or suspicious cytology. In general, the probability of malignancy is considered to be approximately 30% and 70% for equivocal and suspicious cytology, respectively [ 2 , 3 , 4 , 24 ].…”

Section: Discussionmentioning

confidence: 99%

“…Oral potentially malignant disorders (OPMD), including mainly leukoplakia, erythroplakia and oral lichen planus, are mucosal diseases that have a risk of malignancy being present at the time of initial diagnosis or a future date [ 3 ]. The majority of these OPMDs may not progress to OSCC, yet they provide a field of abnormalities in which possible cancer development is more likely than in patients without such disorders [ 4 ].…”

Section: Introductionmentioning

confidence: 99%

Evaluation of Static DNA Ploidy Analysis Using Conventional Brush Biopsy-Based Cytology Samples as an Adjuvant Diagnostic Tool for the Detection of a Malignant Transformation in Potentially Oral Malignant Diseases: A Prospective Study

Bechstedt

Pomjanski

Schramm

et al. 2022

Cancers

View full text Add to dashboard Cite

Background: The accuracy of DNA image cytometry as an investigation method for potentially malignant disorders of the oral cavity is currently still a subject of controversy, due to inconsistently applied definitions of DNA aneuploidy, small cohorts and different application techniques of the method. The aim of this study was to examine the accuracy of the method as a supplementary diagnostic tool in addition to the cytological examination using internationally consented definitions for DNA aneuploidy. Methods: A total of 602 samples from 467 patients with various oral lesions were included in this prospective study. Brush biopsies from each patient were first cytologically examined and categorized by a pathologist, second evaluated using DNA image cytometry, and finally compared to either histological biopsy result or clinical outcome. Results: Using the standard definition of DNA aneuploidy, we achieved a sensitivity of 93.5%, a positive predictive value for the detection of malignant cells of 98.0%, and an area under the curve of 0.96 of DNA ploidy analysis for the detection of severe oral epithelial dysplasia, carcinoma in situ or oral squamous cell carcinoma. Importantly, using logistic regression and a two-step model, we were able to describe the increased association between DNA-ICM and the detection of malignant cells (OR = 201.6) as a secondary predictor in addition to cytology (OR = 11.90). Conclusion: In summary, this study has shown that DNA ploidy analysis based on conventional specimens of oral brush biopsies is a highly sensitive, non-invasive, patient-friendly method that should be considered as an additional diagnostic tool for detecting malignant changes in the oral cavity.

show abstract

Oral potentially malignant disorders: Proceedings from an expert symposium

Cited by 9 publications

References 13 publications

Deep convolutional neural network algorithm for the automatic segmentation of oral potentially malignant disorders and oral cancers

Deep convolutional neural network algorithm for the automatic segmentation of oral potentially malignant disorders and oral cancers

Significance of immunohistochemistry biomarkers in prediction of malignant transformation of oral lichen planus: A systematic review

Evaluation of Static DNA Ploidy Analysis Using Conventional Brush Biopsy-Based Cytology Samples as an Adjuvant Diagnostic Tool for the Detection of a Malignant Transformation in Potentially Oral Malignant Diseases: A Prospective Study

Contact Info

Product

Resources

About