The aim of the present study was the evaluation of the performance of analytical models developed with both mid-infrared (MIR) and near-infrared (NIR) spectral data, to assess the phytochemical composition and in vitro antioxidant activity, besides individual phenolic compounds determined by HPLC-DAD, of flours from 21 distinct cultivars of Phaseolus vulgaris L. Partial least squares (PLS) regression was used to develop the analytical models, which were validated with an external set of samples. In MIR, the best prediction models were developed using the first derivative after normalization (R²c 0.86–0.99 and R²v 0.75–0.94), while for NIR, the use of the first derivative of the spectra after normalization led to the best results (R²c 0.94–0.99 and R²v 0.85–0.97). Both techniques allowed to ascertain the prediction models to ensure an accurate evaluation of the individual phenolic compounds in concentrations as low as ~ 5 μg g⁻¹ and in vitro antioxidant capacity until the lower limit of 2.1 μmol g⁻¹ dw. Therefore, this study revealed that the spectroscopic methodologies may represent an accurate and rapid method for quantification of phytochemical composition, in vitro antioxidant activity and individual phenolic compounds of bean flours; thus, their applicability in the food industry is representing an alternative to the traditional approaches.

本研究的目的是评估使用中红外（MIR）和近红外（NIR）光谱数据开发的分析模型的性能，以评估植物化学成分和体外抗氧化活性，除了确定单个酚类化合物外通过HPLC-DAD对来自菜豆的21个不同品种的面粉进行了分析。使用偏最小二乘（PLS）回归来建立分析模型，并用一组外部样本进行了验证。在MIR中，使用归一化后的一阶导数（R ² c 0.86–0.99和R ² v 0.75–0.94）来开发最佳预测模型，而对于NIR，归一化后使用光谱的一阶导数可以得到最佳的预测模型。结果（R ² c 0.94–0.99和R ² v 0.85–0.97）。两种技术都可以确定预测模型，以确保准确评估浓度低至〜5μgg ^-1的单个酚类化合物和体外抗氧化能力，直到下限为2.1μmolg ^-1  dw。因此，这项研究表明，光谱学方法可以代表一种准确，快速的定量豆粉植物化学成分，体外抗氧化活性和单个酚类化合物的方法。因此，它们在食品工业中的适用性是对传统方法的替代。