Related Experiment Video
Updated: Aug 5, 2026
05:49
Reliability of Artificial Intelligence-Based Cone Beam Computed Tomography Integration with Digital Dental Images
Published on: February 23, 2024
Image-Based Artificial Intelligence for Predicting Malignant Transformation of Oral Potentially Malignant Disorders:
Shaul Hameed Kolarkodi1, Faraj Alotaiby1, Mohammed Fakhry Almutairy1
1Department of Oral and Maxillofacial Diagnostic Sciences, College of Dentistry, Qassim University, Buraydah 52571, Saudi Arabia.
Journal of Clinical Medicine
|July 28, 2026
Summary
Artificial intelligence (AI) shows promise in analyzing images for oral potentially malignant disorders (OPMDs), offering good predictive performance for cancer risk. Further research is needed to standardize methods and validate findings for clinical use.
Area of Science:
- Oral pathology and oncology
- Artificial intelligence in medicine
- Medical imaging analysis
Background:
- Oral potentially malignant disorders (OPMDs) have variable transformation risks to oral squamous cell carcinoma (OSCC).
- Traditional histopathologic grading of dysplasia has poor predictive value and inter-observer variability.
- AI-based image analysis methods are explored for non-invasive risk stratification of OPMDs.
Purpose of the Study:
- To review the current evidence on AI applications in OPMD diagnosis and treatment.
- To assess the methodological characteristics and predictive accuracy of AI methods.
- To identify research priorities for AI in OPMD risk prediction.
Main Methods:
- Scoping review using Joanna Briggs Institute methodology and PRISMA-ScR guidelines.
- Searched PubMed/MEDLINE, Scopus, Web of Science, and Embase (January 2018 - March 2026).
- Included studies using quantitative image analysis, machine learning (ML), or deep learning (DL) for OPMD diagnosis, prognosis, or risk stratification.
Main Results:
- 24 studies (16 primary image-based) met inclusion criteria; most were retrospective.
- Computational pathomics, clinical-photograph deep learning, and optical/spectroscopic imaging were common AI methods.
- AI demonstrated good predictive performance (AUROC 0.73-0.97) but lacked external validation and prospective design in most studies.
Conclusions:
- AI-based image recognition for OPMDs shows strong predictive potential but requires further development.
- Future research should focus on multicenter studies, image standardization, external validation, and improved reporting.
- Radiologic radiomics remains an unexplored area for OPMD risk prediction.
