Clinical Update quiz
The Clinical Update online is a free member service providing up to 11 FREE CPD hours per year.
Each month, the Clinical Update is published online. There are 11 Clinical Updates per year (February - December). This service is available to ADAVB and ADATas members only. Members can log in to view and answer the Clinical Update questions. If you answer at least eight out of 10 questions correctly, you will receive one hour of scientific CPD.
Please note: Each new Clinical Update will be available below from the first business day of each month.
October Clinical Update
The detection of apical radiolucencies in periapical radiographs: A comparison between an artificial intelligence platform and expert endodontists with CBCT serving as the diagnostic benchmark.
Allihaibi M, Koller G, & Mannocci F. (2025). International Endodontic Journal.
Reviewed and edited by Professor Harcourt and Dr Condon. Compiled by Francis Chan.
This article was originally published by Wiley Online Library and has been edited for brevity and clarity.
Introduction
Apical periodontitis is a common chronic inflammatory condition affecting an estimated 52% of individuals globally. Periapical radiographs are widely used for identifying periapical radiolucent lesions (PARLs) associated with apical periodontitis but have limitations that affect accuracy. These include the two-dimensional nature of these images, anatomical noise, and geometric distortion.
Cone beam computed tomography (CBCT) addresses many of these shortcomings. It exhibits higher sensitivity for the detection of apical periodontitis compared with periapical radiographs, using histology as the gold standard. However, its use is limited by higher associated costs, increased radiation exposure, and limited availability. Therefore, there is a need for improved diagnostic accuracy within the scope of 2D imaging modalities.
Artificial intelligence (AI) has emerged as a promising tool to enhance diagnostic accuracy in various medical fields, including dentistry. Commercially available AI-driven platforms (e.g., Diagnocat) use neural networks trained on large datasets to assist in dental diagnosis. These are designed as screening and decision-support tools. Earlier studies show promising results in periapical radiograph analysis but are limited by small sample sizes and reliance on expert opinion of periapical radiographs as the reference standard, rather than using CBCT.
The aim of this study was to:
- Determine the accuracy of Diagnocat for detecting apical radiolucencies on periapical radiographs of untreated teeth, compared with CBCT.
- Compare the performance between Diagnocat and expert clinicians for detecting apical radiolucencies, using CBCT as the reference standard.
Members click here to continue reading and complete this month's Clinical Update questions online or download the current quiz below.
