Related Experiment Video
Updated: Sep 15, 2025
Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
Published on: February 23, 2024
Assessing AI-Based Software's Precision in Identifying Oral Lesions from Radiographs
Hamzah Ali Babkair1, Mohammed Enamur Rashid1, Abedalla Abdelghani1
1Department of Oral and Maxillofacial Diagnostic Sciences, College of Dentistry, Taibah University, Al Madinah Al Munawara, Kingdom of Saudi Arabia.
Background:
Artificial intelligence (AI) is revolutionizing diagnostic practices in dentistry by enhancing accuracy and efficiency. Accurate diagnosis of oral lesions from radiographs is critical for early intervention and treatment planning. This study evaluates the diagnostic accuracy of AI-based software compared to expert radiologists in identifying oral lesions.
Materials And Methods:
A total of 500 radiographic images were collected from a dental teaching hospital. The images included common oral lesions such as cysts, tumors, and infections. AI-based diagnostic software was used to analyze the images, and its performance was compared to that of three experienced radiologists. Sensitivity, specificity, and accuracy were calculated for both methods. Statistical analysis was performed using the Chi-square test, with a significance level set at P < 0.05.
Results:
The AI-based software demonstrated an overall sensitivity of 92%, specificity of 88%, and accuracy of 90%. In comparison, the expert radiologists showed an average sensitivity of 95%, specificity of 91%, and accuracy of 93%. The AI software performed better in detecting small lesions (accuracy: 88%) but was slightly less accurate for complex cases such as mixed radiolucent and radiopaque lesions (accuracy: 86%).
Conclusion:
AI-based diagnostic software is a promising tool for diagnosing oral lesions from radiographs, offering high sensitivity and accuracy. While it performs comparably to expert radiologists in most cases, further optimization is needed for complex lesion types. The integration of AI in routine diagnostic workflows could significantly enhance clinical efficiency and decision-making.
