The use of deep learning state‐of‐the‐art architectures for oral epithelial dysplasia grading: A comparative appraisal
Authors:
- Anna Luíza Damaceno AraújoAnna Luíza Damaceno Araújo
- Viviane Mariano da SilvaVSViviane Mariano da Silva
- Matheus Cardoso MoraesMMMatheus Cardoso Moraes
- Henrique Alves de AmorimHAHenrique Alves de Amorim
- Felipe Paiva FonsecaFFFelipe Paiva Fonseca
- Maria Sissa Pereira Sant'ANaMSMaria Sissa Pereira Sant'ANa
- Ricardo Alves MesquitaRMRicardo Alves Mesquita
- Bruno Augusto Linhares Almeida MarizBMBruno Augusto Linhares Almeida Mariz
- Hélder Antônio Rebelo PontesHPHélder Antônio Rebelo Pontes
- Lucas Lacerda de SouzaLSLucas Lacerda de Souza
- Cristina Saldivia‐SiracusaCSCristina Saldivia‐Siracusa
- Syed Ali KhurramSKSyed Ali Khurram
- 15 September 2023
- journal article
- research article
- Published by Wiley in Journal of Oral Pathology & Medicine
- Vol. 52 (10) , 980-987
Abstract
Background: Dysplasia grading systems for oral epithelial dysplasia are a source of disagreement among pathologists. Therefore, machine learning approaches are being developed to mitigate this issue.Methods: This cross‐sectional study included a cohort of 82 patients with oral potentially malignant disorders and correspondent 98 hematoxylin and eosin‐stained whole slide images with biopsied‐proven dysplasia. All whole‐slide images were manually annotated based on the binary system for oral epithelial dysplasia. The annotated regions of interest were segmented and fragmented into small patches and non‐randomly sampled into training/validation and test subsets. The training/validation data were color augmented, resulting in a total of 81,786 patches for training. The held‐out independent test set enrolled a total of 4,486 patches. Seven state‐of‐the‐art convolutional neural networks were trained, validated, and tested with the same dataset.Results: The models presented a high learning rate, yet very low generalization potential. At the model development, VGG16 performed the best, but with massive overfitting. In the test set, VGG16 presented the best accuracy, sensitivity, specificity, and area under the curve (62%, 62%, 66%, and 65%, respectively), associated with the higher loss among all Convolutional Neural Networks (CNNs) tested. EfficientB0 has comparable metrics and the lowest loss among all convolutional neural networks, being a great candidate for further studies.Conclusion: The models were not able to generalize enough to be applied in real‐life datasets due to an overlapping of features between the two classes (i.e., high risk and low risk of malignization).Keywords
Funding Information
- Fundação de Amparo à Pesquisa do Estado de São Paulo (2021/14585‐7, 2009/53839‐2)
- Coordenação de Aperfeiçoamento de Pessoal de Nível Superior
- Conselho Nacional de Desenvolvimento Científico e Tecnológico
This publication has 23 references indexed in Scilit:
- Interobserver agreement in dysplasia grading: toward an enhanced gold standard for clinical pathology trialsOral Surgery, Oral Medicine, Oral Pathology and Oral Radiology, 2015
- Going deeper with convolutions2015
- Grading of oral epithelial dysplasia: Points to ponderJournal of Oral and Maxillofacial Pathology, 2015
- Automated classification of oral premalignant lesions using image cytometry and Random Forests-based algorithmsCellular Oncology, 2014
- The binary oral dysplasia grading system: validity testing and suggested improvementOral Surgery, Oral Medicine, Oral Pathology and Oral Radiology, 2013
- Texture based segmentation of epithelial layer from oral histological imagesMicron, 2011
- Potentially malignant disorders of the oral and oropharyngeal mucosa; present concepts of managementOral Oncology, 2010
- Automated classification of cells in sub-epithelial connective tissue of oral sub-mucous fibrosis—An SVM based approachComputers in Biology and Medicine, 2009
- Why oral histopathology suffers inter-observer variability on grading oral epithelial dysplasia: An attempt to understand the sources of variationOral Oncology, 2007
- Evaluation of a new binary system of grading oral epithelial dysplasia for prediction of malignant transformationOral Oncology, 2006