Impact of advanced noise reduction algorithms on the diagnostic accuracy of vertical root fractures in cone-beam computed tomography: An evaluation of the influence of intracanal post types

先进降噪算法对锥形束计算机断层扫描诊断垂直根折准确性的影响:根管桩类型的影响评估

阅读:1

Abstract

PURPOSE: Vertical root fractures (VRFs) remain a major diagnostic challenge in endodontics due to image noise and artifacts in cone-beam computed tomography (CBCT), particularly those caused by intracanal posts. This study evaluated the effect of an advanced noise reduction (ANR) algorithm on CBCT performance in detecting VRFs in maxillary second premolars and examined how different post materials influenced diagnostic outcomes. MATERIALS AND METHODS: Seventy extracted maxillary second premolars with single canals were divided into 5 groups according to post material (cast, fiberglass, titanium, stainless steel, and brass). VRFs were induced in half of the specimens using a universal testing machine under standardized conditions. CBCT scans were obtained using a Carestream 9600 device with fixed parameters, both with and without ANR. Two experienced radiologists independently evaluated the images using a 5-point scale. Sensitivity, specificity, predictive values, and interobserver agreement were analyzed using SPSS version 21 (IBM Corp., Armonk, NY, USA). RESULTS: The application of ANR increased overall sensitivity and interobserver agreement compared with conventional images. Specificity varied by post material: fiberglass posts demonstrated the highest diagnostic accuracy, while stainless steel and brass produced stronger artifacts and lower sensitivity. CONCLUSION: Incorporating ANR into CBCT imaging improves VRF detection by improving sensitivity and observer consistency, especially in cases with minimal metallic interference. These findings highlight the clinical benefits of ANR and support further research integrating noise and metal artifact reduction techniques with artificial intelligence to optimize diagnostic precision.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。