A rapid approach for discriminating Ganoderma species using Attenuated Total Reflectance-Fourier Transform Infrared (ATR-FTIR) Spectroscopy integrated with chemometric analysis and Convolutional Neural Network (CNN)

Sze Yun Chen^1,2CHI YUAN LOW²Jun Yang Loh²Wan Yin Tew²Li Yun Ou Yang²Peng Shun Ong²Chong Seng Yan²Hui Wei Loh²Ying Chen¹Wei Xu¹Wen Xu^1*Tiem Leong Yoon^3*Mun Fei Yam^1,2*

• ¹Fujian University of Traditional Chinese Medicine, Fuzhou, China
• ²Universiti Sains Malaysia Pusat Pengajian Sains Farmasi, Minden Heights, Malaysia
• ³Universiti Sains Malaysia Pusat Pengajian Sains Fizik, Minden Heights, Malaysia

ABSTRACT This research addresses the issue of adulteration and misclassification of Ganoderma species. The study presents a novel and comprehensive framework for Ganoderma authentication by analysing Attenuated Total Reflectance-Fourier Transform Infrared (ATR-FTIR) spectra using a combined approach of chemometric analysis and deep learning using Convolutional Neural Network (CNN). The three Ganoderma species involved in this study were Ganoderma lucidum, Ganoderma sinense and Ganoderma tsugae. Among chemometric models, Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA) yielded a high accuracy of 98.61%, a sensitivity of 97.92% and a specificity of 98.96%. Additionally, the RMSEE, RMSEP and RMSECV values for the OPLS-DA model were <0.3, confirming its reliability. The Convolutional Neural Network (CNN) model also performed well, achieving 89.84% accuracy, 84.75% sensitivity, and 92.38% specificity, with minimal variation during random segregation testing. Additionally, the model exhibited a precision of 0.87 ± 0.02, a recall of 0.85 ± 0.03, and an F1 score of 0.86 ± 0.03 for 10 random segregation tests. As a conclusion, both chemometric and CNN models developed in this study are efficient and robust for classifying Ganoderma species. To further validate this combined approach, we aim to implement chemometric and CNN models in other medicinal herbs authentication in the future.