Concentration residual augmented classical least squares (CRACLS) and spectral residual augmented classical least squares (SRACLS) are reflecting the resolving power of spectrophotometric augmented mathematical techniques. The applied multivariate calibration methods were able to determine simultaneously the quinary mixture of nifuroxazide (NIF) and its four carcinogenic impurities, and a comparative study between the proposed models and the conventional partial least squares (PLS) algorithm was also done. Regression models were built covering the range of 10.00 to 50.00 μg mL−1 for NIF, 0.05 to 0.45 μg mL−1 (for each of impurities A and B), and 0.10 to 0.90 μg mL−1 (for impurities C and D). NIF and its four genotoxic impurities were successfully determined in the prepared mixtures and dosage forms. The efficiency of the applied algorithm for resolution and quantitation of the overlapped UV signals were compared, and the advantages over PLS mode were emphasized. The obtained results were statistically compared with each other along with those of the reported method.

中文翻译：

---

两种增强经典最小二乘算法和 PLS 的比较，用于使用紫外光谱测定硝呋噻嗪及其基因毒性杂质

浓度残差增强经典最小二乘法 (CRACLS) 和光谱残差增强经典最小二乘法 (SRACLS) 反映了分光光度法增强数学技术的分辨能力。所应用的多元校准方法能够同时测定硝呋噻嗪 (NIF) 及其四种致癌杂质的五元混合物，并且还对所提出的模型与常规偏最小二乘法 (PLS) 算法进行了比较研究。建立的回归模型涵盖了 NIF 的 10.00 至 50.00 μg mL-1、0.05 至 0.45 μg mL-1（对于每个杂质 A 和 B）和 0.10 到 0.90 μg mL-1（对于杂质 C 和 D） . 在制备的混合物和剂型中成功测定了 NIF 及其四种遗传毒性杂质。比较了应用算法对重叠 UV 信号的分辨率和定量的效率，并强调了与 PLS 模式相比的优势。所得结果与所报告方法的结果进行统计学比较。