Deep learning for detecting periapical bone rarefaction in panoramic radiographs: a systematic review and critical assessment.

OBJECTIVES

To evaluate deep learning (DL)-based models for detecting periapical bone rarefaction (PBRs) in panoramic radiographs (PRs), analysing their feasibility and performance in dental practice.

METHODS

A search was conducted across seven databases and partial grey literature up to November 15, 2024, using Medical Subject Headings and entry terms related to DL, PBRs, and PRs. Studies assessing DL-based models for detecting and classifying PBRs in conventional PRs were included, while those using non-PR imaging or focusing solely on non-PBR lesions were excluded. Two independent reviewers performed screening, data extraction, and quality assessment using the Quality Assessment of Diagnostic Accuracy Studies-2 tool, with conflicts resolved by a third reviewer.

RESULTS

Twelve studies met the inclusion criteria, mostly from Asia (58.3%). The risk of bias was moderate in 10 studies (83.3%) and high in 2 (16.7%). DL models showed moderate to high performance in PBR detection (sensitivity: 26%-100%; specificity: 51%-100%), with U-NET and YOLO being the most used algorithms. Only one study (8.3%) distinguished Periapical Granuloma from Periapical Cysts, revealing a classification gap. Key challenges included limited generalization due to small datasets, anatomical superimpositions in PRs, and variability in reported metrics, compromising models comparison.

CONCLUSION

This review underscores that DL-based has the potential to become a valuable tool in dental image diagnostics, but it cannot yet be considered a definitive practice. Multicentre collaboration is needed to diversify data and democratize those tools. Standardized performance reporting is critical for fair comparability between different models.

ADVANCES IN KNOWLEDGE: This study represents the first critical synthesis on this theme, examining a group of lesions with complex manifestations that have been neglected in comparable technological development studies, where research focus has usually been limited to radicular cysts. We identified gaps in classification tasks, insufficient use of ethnically diverse and heterogeneous datasets, and the need for multicentric studies. The variability in data reporting prevents transparent comparisons, even precluding our planned meta-analysis. Consequently, we emphasize the necessity for standardized reporting protocols similar to PRISMA for systematic reviews or STARD for diagnostic or prognostic studies, particularly since accuracy metrics remain inadequately documented while critically important.