Torabinejad Dental Research Center, Department of Oral and Maxillofacial Radiology, School of Dentistry, Isfahan University of Medical Sciences, Isfahan, Iran.
Dental Implant Research Center, Department of Periodontics, Dental Research Institute, Isfahan University of Medical Sciences, Isfahan, Iran.
J Periodontol. 2020 Feb;91(2):209-214. doi: 10.1002/JPER.18-0433. Epub 2019 Aug 28.
The aim of the present study was to investigate the impact of metal artifact reduction (MAR) algorithm of cone-beam computed tomography (CBCT) on the diagnostic accuracy of fenestration and dehiscence around dental implants.
Twenty-three dental implants were placed adjacent to the dehiscence and 23 adjacent to the fenestration defects on bovine bone blocks. The blocks were scanned with CBCT unit in two modes, with and without MAR algorithm. The area under the receiver operator characteristic (ROC) curves (Az value), specificity, sensitivity, positive predictive value, negative predictive value, and accuracy were determined for all modes.
For both defects, the Az values were higher in off MAR condition. The values of sensitivity, positive predictive value, negative predictive value, and accuracy, were higher in off MAR condition for both defects. However, the specificity in both defects in the two modes was equal.
The MAR algorithm in CBCT unit may not be helpful in increasing the diagnostic accuracy of fenestration and dehiscence around dental implants.
本研究旨在探讨锥形束计算机断层扫描（CBCT）的金属伪影降低（MAR）算法对种植体周围开窗和骨开裂诊断准确性的影响。
在牛骨块上，将 23 颗种植体分别放置在骨开裂和开窗缺陷的邻接处。使用 CBCT 单元以两种模式（有和无 MAR 算法）对这些骨块进行扫描。确定所有模式的接收者操作特征（ROC）曲线下面积（Az 值）、特异性、敏感性、阳性预测值、阴性预测值和准确性。
对于两种缺陷，在关闭 MAR 算法的情况下，Az 值更高。在两种缺陷中，关闭 MAR 算法时，敏感性、阳性预测值、阴性预测值和准确性更高。然而，两种模式下两种缺陷的特异性是相等的。
CBCT 单元中的 MAR 算法可能无助于提高种植体周围开窗和骨开裂的诊断准确性。
