Research Article
BibTex RIS Cite

Oral Cancer Classification with CNN Based State-of-the-art Transfer Learning Methods

Year 2025, Volume: 8 Issue: 1, 17 - 18
https://doi.org/10.34248/bsengineering.1528581

Abstract

Ağız ve diş sağlığının önemi diğer hayati organları da yakından etkilemektedir. Bu çalışmada, iyi huylu ve kötü huylu lezyonlara sahip histopatolojik ve ağız içi görüntüler üzerinde CNN tabanlı transfer öğrenme modelleri oluşturulmuştur. İki farklı kaynaktan alınan histopatolojik ve intraoral görüntüler, ağızdaki lezyonların iyi huylu veya kötü huylu sınıflarına sahiptir. Transfer öğrenme yöntemi olarak EfficientNetB7, ResNet50, VGG16 ve VGG19, Xception, ConvNextBase ve MobileNetV2 kullanılmıştır. Model eğitimi, eğitim seti üzerinde %80-%20 eğitim testi ayrımı ve %20 doğrulama ayrımı ile gerçekleştirilmiştir. Modeli değerlendirmek için Acc, Prec, Rec ve F1 metrikleri kullanılmıştır. Histopatolojik görüntülerde Resnet50 %81,25 Acc ve %85,25 F1 ile öndeydi. Ağız içi görüntülerde, ConvNextBase %0,80 Acc ve %0,80 F1 ile daha doğru bulundu.

Ethical Statement

Etik beyan gerektirecek bir çalışma değildir.

Supporting Institution

Yok

Project Number

None

Thanks

yok

References

  • Babu PA, Rai AK, Ramesh JVN, Nithyasri A, Sangeetha S, Kshirsagar PR, Rajendran A, Rajaram A, Dilipkumar S. 2024. An explainable deep learning approach for oral cancer detection. J Electr Eng Technol., 19: 1837–1848.
  • Bakare YB, Kumarasamy M, 2021. Histopathologıcal image analysis for oral cancer classification by support vector machine. Int J Adv Signal Image Sci, 7: 1–10.

Oral Cancer Classification with CNN Based State-of-the-art Transfer Learning Methods

Year 2025, Volume: 8 Issue: 1, 17 - 18
https://doi.org/10.34248/bsengineering.1528581

Abstract

The importance of oral and dental health closely affects other vital organs. In this study, CNN-based transfer learning models are built on histopathologic and intraoral images with benign and malignant lesions. Histopathologic and intraoral images from two different sources have benign or malignant classes of lesions in the mouth. EfficientNetB7, ResNet50, VGG16, and VGG19, Xception, ConvNextBase, and MobileNetV2 were used as transfer learning methods. Model training was performed with 80%-20% train test separation and 20% validation separation on the train set. Acc, Prec, Rec, and F1 metrics were used to evaluate the model. In histopathologocial images, Resnet50 was ahead with 81.25% Acc and 85.25% F1. In intraoral images, ConvNextBase with 0.80% Acc, and 0.80% F1 was found to be more accurate.

Project Number

None

References

  • Babu PA, Rai AK, Ramesh JVN, Nithyasri A, Sangeetha S, Kshirsagar PR, Rajendran A, Rajaram A, Dilipkumar S. 2024. An explainable deep learning approach for oral cancer detection. J Electr Eng Technol., 19: 1837–1848.
  • Bakare YB, Kumarasamy M, 2021. Histopathologıcal image analysis for oral cancer classification by support vector machine. Int J Adv Signal Image Sci, 7: 1–10.
There are 2 citations in total.

Details

Primary Language English
Subjects Biomedical Imaging
Journal Section Research Articles
Authors

Kaan Gümele 0009-0002-4262-0585

Muhammet Sinan Başarslan 0000-0002-7996-9169

Project Number None
Publication Date
Submission Date August 5, 2024
Acceptance Date November 25, 2024
Published in Issue Year 2025 Volume: 8 Issue: 1

Cite

APA Gümele, K., & Başarslan, M. S. (n.d.). Oral Cancer Classification with CNN Based State-of-the-art Transfer Learning Methods. Black Sea Journal of Engineering and Science, 8(1), 17-18. https://doi.org/10.34248/bsengineering.1528581
AMA Gümele K, Başarslan MS. Oral Cancer Classification with CNN Based State-of-the-art Transfer Learning Methods. BSJ Eng. Sci. 8(1):17-18. doi:10.34248/bsengineering.1528581
Chicago Gümele, Kaan, and Muhammet Sinan Başarslan. “Oral Cancer Classification With CNN Based State-of-the-Art Transfer Learning Methods”. Black Sea Journal of Engineering and Science 8, no. 1 n.d.: 17-18. https://doi.org/10.34248/bsengineering.1528581.
EndNote Gümele K, Başarslan MS Oral Cancer Classification with CNN Based State-of-the-art Transfer Learning Methods. Black Sea Journal of Engineering and Science 8 1 17–18.
IEEE K. Gümele and M. S. Başarslan, “Oral Cancer Classification with CNN Based State-of-the-art Transfer Learning Methods”, BSJ Eng. Sci., vol. 8, no. 1, pp. 17–18, doi: 10.34248/bsengineering.1528581.
ISNAD Gümele, Kaan - Başarslan, Muhammet Sinan. “Oral Cancer Classification With CNN Based State-of-the-Art Transfer Learning Methods”. Black Sea Journal of Engineering and Science 8/1 (n.d.), 17-18. https://doi.org/10.34248/bsengineering.1528581.
JAMA Gümele K, Başarslan MS. Oral Cancer Classification with CNN Based State-of-the-art Transfer Learning Methods. BSJ Eng. Sci.;8:17–18.
MLA Gümele, Kaan and Muhammet Sinan Başarslan. “Oral Cancer Classification With CNN Based State-of-the-Art Transfer Learning Methods”. Black Sea Journal of Engineering and Science, vol. 8, no. 1, pp. 17-18, doi:10.34248/bsengineering.1528581.
Vancouver Gümele K, Başarslan MS. Oral Cancer Classification with CNN Based State-of-the-art Transfer Learning Methods. BSJ Eng. Sci. 8(1):17-8.

                                                24890