Approach on the one-step quantitative determination of icariin, epimedin A, epimedin B, and epimedin C in epimedium using handheld near-infrared spectrometry assisted by machine learning algorithms

This study aims to expand the application of handheld NIR combined with machine learning in quantitative analysis of Epimedium. After establishing and evaluating several promising models (GBR, ET, eXGB, SVR, RF, PLSR), we selected PLSR and SVR for further optimization due to their excellent performance in terms of R2. Employing the NIPPY library for data preprocessing and a genetic algorithm (GA) for feature selection, models for predicting the contents of four active components in Epimedium were successfully established. The GA+SVR model achieved R2 values of 0.9556 (icariin), 0.9586 (epimedin A), 0.9432 (epimedin B), and 0.9232 (epimedin C), with RMSE values as low as 0.0031 (icariin), 0.0004 (epimedin A), 0.0003 (epimedin B), and 0.0182 (epimedin C). These findings establish a novel paradigm for high-throughput analysis and quality assessment of food and TCM active ingredients.