当前位置:
X-MOL 学术
›
Vib. Spectrosc.
›
论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Application of near infrared spectroscopy with chemometrics for qualitative and quantitative dental caries assessment
Vibrational Spectroscopy ( IF 2.5 ) Pub Date : 2020-11-01 , DOI: 10.1016/j.vibspec.2020.103170 Patteera Sodata , Jomjai Peerapattana
Vibrational Spectroscopy ( IF 2.5 ) Pub Date : 2020-11-01 , DOI: 10.1016/j.vibspec.2020.103170 Patteera Sodata , Jomjai Peerapattana
Abstract This study aimed to apply near infrared spectroscopy with chemometrics to differentiate various stages of teeth, as well as to predict the enamel surface changes and the crystallite size of the carious specimens. Qualitative analysis based on principal component analysis revealed that spectral data pretreated with Savitzky-Golay derivative followed by standard normal variate could differentiate sound teeth, artificial carious teeth, and natural carious teeth with sufficient accuracy. The first principal component described 98.82 % of the variability in the spectral data. With respect to quantitative analysis, the carious specimens were divided to five groups; non-treatment and treatment with various pharmaceutical preparations. The calibration model was constructed from the spectral data training set by a partial least square regression. As a result, the model based on spectral data pretreated by the first derivative method could predict the enamel surface changes and the crystallite size accurately. The model had the lowest root mean square error of prediction, the highest correlation coefficient and acceptable bias value. This study demonstrated that near infrared spectroscopy was a convenient and reliable tool for dental caries assessment.
中文翻译:
近红外光谱与化学计量学在龋齿定性和定量评估中的应用
摘要 本研究旨在应用近红外光谱结合化学计量学来区分牙齿的各个阶段,并预测龋齿标本的牙釉质表面变化和微晶尺寸。基于主成分分析的定性分析表明,用Savitzky-Golay导数和标准正态变量预处理的光谱数据可以足够准确地区分健全牙齿、人工龋齿和天然龋齿。第一个主成分描述了光谱数据中 98.82% 的可变性。在定量分析方面,龋标本分为五组;不治疗和用各种药物制剂治疗。校准模型是通过偏最小二乘回归从光谱数据训练集构建的。因此,基于经一阶导数法预处理的光谱数据的模型可以准确预测釉质表面变化和微晶尺寸。该模型具有最低的预测均方根误差、最高的相关系数和可接受的偏差值。这项研究表明,近红外光谱是一种方便可靠的龋齿评估工具。
更新日期:2020-11-01
中文翻译:
近红外光谱与化学计量学在龋齿定性和定量评估中的应用
摘要 本研究旨在应用近红外光谱结合化学计量学来区分牙齿的各个阶段,并预测龋齿标本的牙釉质表面变化和微晶尺寸。基于主成分分析的定性分析表明,用Savitzky-Golay导数和标准正态变量预处理的光谱数据可以足够准确地区分健全牙齿、人工龋齿和天然龋齿。第一个主成分描述了光谱数据中 98.82% 的可变性。在定量分析方面,龋标本分为五组;不治疗和用各种药物制剂治疗。校准模型是通过偏最小二乘回归从光谱数据训练集构建的。因此,基于经一阶导数法预处理的光谱数据的模型可以准确预测釉质表面变化和微晶尺寸。该模型具有最低的预测均方根误差、最高的相关系数和可接受的偏差值。这项研究表明,近红外光谱是一种方便可靠的龋齿评估工具。