目的：建立与验证快速分析双黄连口服液中黄芩苷含量的近红外光谱法。方法：采用透射模式采集样品近红外光谱数据，以偏最小二乘法建立光谱数据与HPLC 法分析结果的定量校正模型，根据R、RMSEC、RMSEP 等参数评价模型的定量性能；以准确性轮廓方法，验证近红外模型的准确度和精密度。结果：偏最小二乘定量校正模型R 为0.9902，RMSEC 为0.4223，RMSEP 为0.4162；在茁-期望容许误差极限为95%、容许极限为12%、风险极限为5%的约束下，模型验证结果表明分析方法准确度、精密度、不确定度满足要求，最低定量限为5.32 mg·mL-1。结论：双黄连口服液中黄芩苷近红外定量模型稳健可靠。此外，准确性轮廓作为一种新的验证方法，可运用于中药体系近红外分析模型验证。

This study was aimed to develop and validate a near-infrared (NIR) method to accurately determine baicalin in the Shuanghuanglian oral liquid. Transmission mode was used in the collection of NIR spectral samples. A robust PLS model was established based on the correlation between the NIR spectrum and HPLC reference. And quantitative property of the established model was evaluated through indexes of R, RMSEC, and RMSEP. The accuracy profile was introduced as a new method to validate the accuracy and trueness of the model. The results showed that conventional criteria such as the R, RMSEC, and RMSEP are 0.9902, 0.4223 and 0.4162, respectively. The limit of quality (LOQ) of the model is from 5.324 mg/mL to 11.14 mg/mL when the β-expectations tolerance limits, accept limit and risk profile was at 95%, 12% and 5%, respectively. The results were with high truness, intermediate precision and recovery. It was concluded that the NIR quantitative model of baicalin in Shuanghuanglian oral liquid is reliable. In addition, accuracy profile as a new validation method which can be used to evaluate NIR model in traditional Chinese medicine.

科学技术部国家“重大新药创制”科技重大专项（2010ZX09502-002）：清开灵注射液安全性关键技术研究，负责人：乔延江。