Artificial intelligence vs. semi-automated segmentation for assessment of dental periapical lesion volume index score: A cone-beam CT study.

INTRODUCTION: Cone beam computed tomography periapical volume index (CBCTPAVI) is a categorisation tool to assess periapical lesion size in three-dimensions and predict treatment outcomes. This index was determined using a time-consuming semi-automatic segmentation technique. This study compared artificial intelligence (AI) with semi-automated segmentation to determine AI's ability to accurately determine CBCTPAVI score. METHODS: CBCTPAVI scores for 500 tooth roots were determined using both the semi-automatic segmentation technique in three-dimensional imaging analysis software (Mimics Research™) and AI (Diagnocat™). A confusion matrix was created to compare the CBCTPAVI score by the AI with the semi-automatic segmentation technique. Evaluation metrics, precision, recall, F1-score (2×precision×recallprecision+recall), and overall accuracy were determined. RESULTS: In 84.4 % (n = 422) of cases the AI classified CBCTPAVI score the same as the semi-automated technique. AI was unable to classify any lesion as index 1 or 2, due to its limitation in small volume measurement. When lesions classified as index 1 and 2 by the semi-automatic segmentation technique were excluded, the AI demonstrated levels of precision, recall and F1-score, all above 0.85, for indices 0, 3-6; and accuracy over 90 %. CONCLUSIONS: Diagnocat™ with its ability to determine CBCTPAVI score in approximately 2 min following upload of the CBCT could be an excellent and efficient tool to facilitate better monitoring and assessment of periapical lesions in everyday clinical practice and/or radiographic reporting. However, to assess three-dimensional healing of smaller lesions (with scores 1 and 2), further advancements in AI technologies are needed.

Sphericity of Periapical Lesion and Its Relation to the Novel CBCT Periapical Volume Index.

BACKGROUND: The purpose of this study was to assess the sphericity of periapical lesions and its relation to the cone-beam computed tomography periapical volume index (CBCTPAVI). METHODS: A total of 261 periapical lesions were assessed using cone-beam computed tomography images from InteleViewer. Three-dimensional analysis of the lesions was conducted using analytical imaging software Mimics Research. Lesion volume and surface area measurements were determined using the semiautomatic segmentation technique, and these measurements were then used to determine lesion sphericity and CBCTPAVI score. One-way analysis of variance with post hoc Tukey test was used to assess for differences in sphericity among CBCTPAVI groups. RESULTS: The mean sphericity of periapical lesions was 62%. Periapical lesions with larger CBCTPAVI scores were significantly less spherical than lesions with smaller CBCTPAVI scores. CONCLUSIONS: Periapical lesions of endodontic origin are mostly semi-spherical in their spread, and as CBCTPAVI score increases, sphericity decreases, indicating that larger lesions expand less uniformly compared with smaller lesions. Clinicians should be aware that lesions of increased volume have less sphericity, and are thus elongated or stretched in 1 or more anatomic plane. This information will assist clinicians in planning and performing periapical surgery and may aid in differential diagnosis of radiolucent jaw lesions.

A Novel Volume-based Cone-beam Computed Tomographic Periapical Index.

INTRODUCTION: The purpose of this study was to evaluate the variations in the volume of periapical lesions scored using a cone-beam computed tomographic periapical index (CBCTPAI) and to develop a new volume-based periapical index. METHODS: Cone-beam computed tomographic images were obtained from InteleViewer (Intelerad Medical Systems Incorporated, Montreal, Canada). Teeth with a periapical radiolucency or with a history of endodontic treatment were included in this study. Using 3-dimensional medical imaging processing software (Mimics Research; Materialise NV, Leuven, Belgium), the maximum diameter of 273 periapical lesions and their corresponding CBCTPAI score was determined. The software was then used to determine the volume of the lesions using a semiautomatic segmentation technique. RESULTS: There was a substantial variation in the volume for CBCTPAI scores 3, 4, and 5, which was demonstrated by the variance and range, thus making it difficult to use the current CBCTPAI as a method to predict volume and treatment outcomes. A new index, the cone-beam computed tomographic periapical volume index (CBCTPAVI), was developed using partition classification analysis. The results for the new index demonstrated high levels of sensitivity, specificity, precision, and area under the curve, all at 0.90 or more, except 1 sensitivity for CBCTPAVI 1 at 0.875. Overall, the accurate classification rate was 98.169%, and the root mean square error rate was low at 0.07. CONCLUSIONS: The proposed CBCTPAVI will allow clinicians to classify lesions based on their true 3-dimensional size, accurately assess healing of lesions, and predict treatment outcomes.