Behere Rohit R, Lele Shailesh M
Department of Oral Medicine and Radiology, Sinhgad Dental College and Hospital, Pune, India.
Indian J Dent Res. 2011 Mar-Apr;22(2):362. doi: 10.4103/0970-9290.84277.
Digital radiography has so far not resulted in improved rates of proximal caries detection. Historically, automated caries detection tools have been largely academic. Opinions regarding the performance of the only such commercially available tool, viz., Logicon caries Detector (LCD) have been equivocal. This study was conducted to evaluate the reliability of LCD in the detection and depth assessment of proximal caries.
Digital images were obtained of 100 proximal tooth surfaces using the Kodak RVG 5000 sensor and analyzed by three observers. The images were then analyzed by the principal investigator using the LCD software. The teeth were then sectioned and magnified photographic images were obtained which were taken as the gold standard. All the grades were entered in proformas and the data were statistically analyzed using the chi-square test. Five parameters of reliability were calculated.
The sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy of LCD for the grade No caries were 33, 96, 73, 82, and 81%, respectively; for the grade Enamel caries were 5, 97, 33, 80, and 79%, respectively; and for the grade Dentin caries were 100, 96, 50, 100, and 96%, respectively.
In conclusion, LCD appears to be more reliable in ruling out (both enamel and dentin) caries than in detecting caries.
迄今为止,数字放射成像技术尚未提高邻面龋的检出率。从历史上看,自动化龋病检测工具大多处于学术研究阶段。对于唯一一款此类商用工具,即Logicon龋病检测仪(LCD)的性能评价一直存在分歧。本研究旨在评估LCD在邻面龋检测及深度评估中的可靠性。
使用柯达RVG 5000传感器获取100个邻面牙齿表面的数字图像,并由三名观察者进行分析。然后由主要研究者使用LCD软件对图像进行分析。随后将牙齿切片并获取放大的照片图像,将其作为金标准。所有分级均记录在表格中,并使用卡方检验对数据进行统计学分析。计算了五个可靠性参数。
LCD对无龋等级的敏感度、特异度、阳性预测值、阴性预测值及总体准确率分别为33%、96%、73%、82%和81%;对釉质龋等级分别为5%、97%、33%、80%和79%;对牙本质龋等级分别为100%、96%、50%、100%和96%。
总之,LCD在排除(釉质和牙本质)龋方面似乎比检测龋更可靠。