Evaluation of Dental Panoramic Radiographs by Artificial Intelligence Compared to Human Reference: A Diagnostic Accuracy Study

Abstract

Background/Objectives: The role of artificial intelligence (AI) in dentistry is becoming increasingly significant, particularly in diagnosis and treatment planning. This study aimed to assess the sensitivity, specificity, accuracy, and precision of AI-driven software in analyzing dental panoramic radiographs (DPRs) in patients with permanent dentition. Methods: Out of 638 DPRs, 600 fulfilled the inclusion criteria. The radiographs were analyzed by AI software and two researchers. The following variables were assessed: (1) missing tooth, (2) root canal filling, (3) endodontic lesion, (4) implant, (5) abutment, (6) pontic, (7) crown, (8) and sound tooth. Results: The study revealed very high performance metrics for the AI algorithm in detecting missing teeth, root canal fillings, and implant abutment crowns, all greater than 90%. However, it demonstrated moderate sensitivity and precision in identifying endodontic lesions and the lowest precision (65.30%) in detecting crowns. Conclusions: AI software can be a valuable tool in clinical practice for diagnosis and treatment planning but may require additional verification by clinicians, especially for identifying endodontic lesions and crowns. Due to some limitations of the study, further research is recommended.

Keywords: artificial intelligence; automatic detection; diagnosis; panoramic radiograph.

Figure 1

Flow diagram.

Figure 2

Number of missing teeth detected by AI algorithm and human examiner, with positional detail in upper (18–28) and lower (38–48) dental arch. Position numbering according to ISO 3950:2016 FDI notation (version 2022, FDI World Dental Federation, Geneva, Switzerland) [31].

Figure 3

Number of root canal fillings and endodontic lesions detected by AI algorithm and human examiner, with positional detail in upper (18–28) and lower (38–48) dental arch. Position numbering according to ISO 3950:2016 FDI notation (version 2022, FDI World Dental Federation, Geneva, Switzerland) [31].

Figure 4

Number of implants, implant abutment crowns, pontic crowns, and dental abutment crowns detected by AI algorithm and human examiner, with positional detail in upper (18–28) and lower (38–48) dental arch. Position numbering according to ISO 3950:2016 FDI notation (version 2022, FDI World Dental Federation, Geneva, Switzerland) [31].

Figure 5

ROC curve representations for the studied variables. The random classifier (diagonal line) is orange, and the ROC curve for the classifier is blue.

Figure 6

Examples of teeth misdiagnosed by AI software. (A) Tooth 26 with an orthodontic band, mistakenly identified as a tooth with a crown. (B) Tooth 36, not treated endodontically, incorrectly diagnosed as a tooth with root canal filling. (C) Teeth 44 and 45 with extensive, highly radiopaque dental fillings, wrongly identified as teeth with crowns. (D) Tooth 44 with incomplete root development, misdiagnosed as a tooth with a periapical lesion.