Development and Validation of a Portable AOTF-NIR Measurement Method for the Determination of Baicalin in Yinhuang Oral Solution

The present study was to develop and validate a hand-held AOTF-NIR measurement method enable to accurately determine low dose of Chinese Herbal Medicines (CHM). Yinhuang oral solutions were taken as an example, and an experimental protocol was then followed, involving two operators and independent production lots for data acquisition. On the basis of this protocol, partial least squares regression (PLS) was then carried out. This result in a calibration with standard errors of calibration (SEC) and determination coefficient (R2) was equal to 146.6 µg/mL and 0.9945. In prediction step, prediction set gave standard errors of prediction (SEP) and R2 of 87.7 µg/mL and 0.9923, respectively. Furthermore, a novel approach based on accuracy profile was used to validate the PLS model. The resulting accuracy profile clearly showed that PLS model was able to determine baicalin content, which the LLOQ was 2020 µg/mL. It was concluded that the hand-held AOTF-NIR measurement method could be used for complicated low content CHM, paving the way for portable environment.