Objective: n this study, in order to test the usability of artificial intelligence technologies in dentistry, which are becoming widespread and expanding day by day, and to investigate ways to benefit more from artificial intelligence technologies; a tooth detection and numbering study was performed on panoramic radiographs using a deep learning software.
 Methods: A radiographic dataset containing 200 anonymous panoramic radiographs collected from individuals over the age of 18 was assessed in this retrospective investigation. The images were separated into three groups: training (80%), validation (10%), and test (10%), and tooth numbering was performed with the DCNN artificial intelligence software.
 Results: The D-CNN system has been successful in detecting and numbering teeth. of teeth. The predicted precision, sensitivity, and F1 score were 0.996 (98.0%), 0.980 (98.0%), and 0.988 (98.8%), respectively.
 Conclusion: The precision, sensitivity and F1 scores obtained in our study were found to be high, as 0.996 (98.0%), 0.980 (98.0%) and 0.988 (98.8%), respectively. Although the current algorithm based on Faster R-CNN shows promising results, future studies should be done by increasing the number of data for better tooth detection and numbering results.