Quality Control of Jati Belanda Leaves (Guazuma ulmifolia) using Image Analysis and Chemometrics

The quality of medicinal plants, such as Guazuma ulmifolia (jati belanda, JB), affects the quality of the herbal material derived from them, and can be determined using image analysis. The objective of this study is to investigate the possibility of using an image-generated spectrum and chemometrics as a method for quality control of Jati belanda leaves. Three different quality levels of JB leaves were determined, based on their harvesting time, and confirmed by total flavonoid content analysis. The images of JB samples were collected and reconstructed as a reflection spectrum using the Wiener estimation. The reconstructed spectrum had a goodness-of-fit coefficient of 0.9576 and a root-mean-square-error (RMSE) of 36.65%, compared to the experimental spectrum. Principal Component Analysis (PCA) was used to classify the JB reconstructed spectrum based on its quality. A score plot of two PCs that represented 98% variance was able to group the JB spectrum. Further analysis using Partial Least Squares-Discriminant Analysis (PLSDA) showed that the method can result in around 90% prediction success rate with external validation. This study indicates that image analysis and chemometrics could be used as quality control methods for herbal material.

Quality Control of Jati Belanda Leaves (Guazuma ulmifolia) using Image Analysis and Chemometrics

The quality of medicinal plants, such as Guazuma ulmifolia (jati belanda, JB), affects the quality of the herbal material derived from them, and can be determined using image analysis. The objective of this study is to investigate the possibility of using an image-generated spectrum and chemometrics as a method for quality control of Jati belanda leaves. Three different quality levels of JB leaves were determined, based on their harvesting time, and confirmed by total flavonoid content analysis. The images of JB samples were collected and reconstructed as a reflection spectrum using the Wiener estimation. The reconstructed spectrum had a goodness-of-fit coefficient of 0.9576 and a root-mean-square-error (RMSE) of 36.65%, compared to the experimental spectrum. Principal Component Analysis (PCA) was used to classify the JB reconstructed spectrum based on its quality. A score plot of two PCs that represented 98% variance was able to group the JB spectrum. Further analysis using Partial Least Squares-Discriminant Analysis (PLSDA) showed that the method can result in around 90% prediction success rate with external validation. This study indicates that image analysis and chemometrics could be used as quality control methods for herbal material.

A High-Throughput Imaging Spectrometer Based on Over-Scanning

We present a novel high-throughput imaging spectrometer based on over-scanning. The traditional slit-based spectrometer cannot gather enough radiation while the source is too weak. A much wider slit is used to replace the narrow one in traditional spectrometer. Too much wide slit will cause overlapping between different wavelength lights. In order to reconstruct super-resolution spectrum of source, over-scanning is employed which is realized by high precision electromechanical device. Experiments show that the reconstructed spectrum achieved a much higher resolution than original data meanwhile the throughput has improved significantly.

Computer Program for Deconvoluting Powder Diffraction Spectra

A program is presented that removes broadening from X-ray diffraction spectra. An instrumental spectrum can be used to describe empirically the broadening to be removed, or a Gaussian, Cauchy, or Pearson-VII distribution can be used to analytically describe it. In either case, singlet or doublet forms can be generated. The program returns the deconvoluted spectrum, the reconstructed spectrum, and a sum-of-squares difference between the original and reconstructed spectra. Deconvolution is accomplished using a combination of least-squares, background, and smoothing criteria that minimizes the effect of random counting errors.