Item request has been placed! ×
Item request cannot be made. ×
loading  Processing Request

Artificial intelligence based vision transformer application for grading histopathological images of oral epithelial dysplasia: a step towards AI-driven diagnosis.

Item request has been placed! ×
Item request cannot be made. ×
loading   Processing Request
Read Online Read More Add to Saved list
  • Additional Information
    • Source:
      Publisher: BioMed Central Country of Publication: England NLM ID: 100967800 Publication Model: Electronic Cited Medium: Internet ISSN: 1471-2407 (Electronic) Linking ISSN: 14712407 NLM ISO Abbreviation: BMC Cancer Subsets: MEDLINE
    • Publication Information:
      Original Publication: London : BioMed Central, [2001-
    • Subject Terms:
      Artificial Intelligence* ; Mouth Neoplasms*/pathology ; Mouth Neoplasms*/diagnosis ; Mouth Mucosa*/pathology ; Image Processing, Computer-Assisted*/methods ; Precancerous Conditions*/pathology; Humans ; Neural Networks, Computer ; Deep Learning ; Neoplasm Grading/methods ; Algorithms ; Image Interpretation, Computer-Assisted/methods
    • Abstract:
      Competing Interests: Declarations. Ethics approval and consent to participate: The Research Ethics Committee of Baqiyatallah University of Medical Sciences (IR.BMSU.REC.1402.076) approved the study design. Consent for publication: Not applicable. Competing interests: The authors declare no competing interests.
      Background: This study aimed to classify dysplastic and healthy oral epithelial histopathological images, according to WHO and binary grading systems, using the Vision Transformer (ViT) deep learning algorithm-a state-of-the-art Artificial Intelligence (AI) approach and compare it with established Convolutional Neural Network models (VGG16 and ConvNet).
      Methods: A total of 218 histopathological slide images were collected from the Department of Oral and Maxillofacial Pathology at Tehran University of Medical Sciences archive and combined with two online databases. Two oral pathologists independently labeled the images based on the 2022 World Health Organization (WHO) grading system (mild, moderate and severe), the binary grading system (low risk and high risk), including an additional normal tissue class. After preprocessing, the images were fed to the ViT, VGG16 and ConvNet models.
      Results: Image preprocessing yielded 2,545 low-risk, 2,054 high-risk, 726 mild, 831 moderate, 449 severe, and 937 normal tissue patches. The proposed ViT model outperformed both CNNs with the accuracy of 94% (VGG16:86% and ConvNet: 88%) in 3-class scenario and 97% (VGG16:79% and ConvNet: 88%) in 4-class scenario.
      Conclusions: The ViT model successfully classified oral epithelial dysplastic tissues with a high accuracy, paving the way for AI to serve as an adjunct or independent tool alongside oral and maxillofacial pathologists for detecting and grading oral epithelial dysplasia.
      (© 2025. The Author(s).)
    • References:
      Med Image Anal. 2023 Aug;88:102802. (PMID: 37315483)
      Dent Res J (Isfahan). 2017 Jul-Aug;14(4):297-298. (PMID: 28928786)
      Radiol Artif Intell. 2024 Jul;6(4):e240300. (PMID: 38809149)
      Oral Dis. 2021 Nov;27(8):1947-1976. (PMID: 34418233)
      Asian Pac J Cancer Prev. 2017 Dec 29;18(12):3251-3254. (PMID: 29286215)
      Med Oral Patol Oral Cir Bucal. 2012 Nov 01;17(6):e935-42. (PMID: 22549675)
      Insights Imaging. 2023 Jan 16;14(1):11. (PMID: 36645542)
      Oral Oncol. 2020 Nov;110:104885. (PMID: 32674040)
      Cell Oncol (Dordr). 2014 Jun;37(3):193-202. (PMID: 24817187)
      Oral Oncol. 2022 Nov;134:106109. (PMID: 36126604)
      J Pathol Inform. 2016 Jul 26;7:29. (PMID: 27563488)
      Dent Clin North Am. 2018 Jan;62(1):1-27. (PMID: 29126487)
      BMC Oral Health. 2024 Apr 9;24(1):434. (PMID: 38594651)
      Cancer Discov. 2021 Apr;11(4):900-915. (PMID: 33811123)
      Head Neck. 2020 Mar;42(3):539-555. (PMID: 31803979)
      Eur Arch Otorhinolaryngol. 2020 Nov;277(11):2967-2976. (PMID: 32447493)
      Front Bioeng Biotechnol. 2019 Nov 01;7:300. (PMID: 31737619)
      Radiol Phys Technol. 2017 Sep;10(3):257-273. (PMID: 28689314)
      Data Brief. 2023 Apr 07;48:109128. (PMID: 37122923)
      Oral Surg Oral Med Oral Pathol Oral Radiol. 2013 Jan;115(1):87-94. (PMID: 23217539)
      Med Oral Patol Oral Cir Bucal. 2017 Sep 1;22(5):e520-e526. (PMID: 28809365)
      J Microsc. 2018 Jan;269(1):85-93. (PMID: 28768053)
      J Am Acad Dermatol. 2019 Jul;81(1):59-71. (PMID: 30447325)
      J Oral Maxillofac Surg. 1990 Nov;48(11):1201-5. (PMID: 2170606)
      J Oral Maxillofac Pathol. 2015 May-Aug;19(2):198-204. (PMID: 26604497)
      Biomed Res Int. 2021 Jun 22;2021:9751564. (PMID: 34258283)
      Head Neck Pathol. 2022 Mar;16(1):54-62. (PMID: 35312982)
      J Big Data. 2021;8(1):53. (PMID: 33816053)
    • Contributed Indexing:
      Keywords: Artificial Intelligence; Deep learning; Histopathological images; Oral epithelial dysplasia
    • Publication Date:
      Date Created: 20250425 Date Completed: 20250426 Latest Revision: 20250428
    • Publication Date:
      20250428
    • Accession Number:
      PMC12032748
    • Accession Number:
      10.1186/s12885-025-14193-x
    • Accession Number:
      40281456