Evaluation of Aspergillus flavus growth and detection of aflatoxin B1 content on maize agar culture medium using Vis/NIR hyperspectral imaging

Authors: GUO, XIAOHUAN - China Agricultural University; JIA, BEIBEI - Chinese Academy Of Inspection And Quarantine; ZHANG, HAICHENG - China Agricultural University; Ni, Xinzhi; Zhuang, Hong; LU, YAO - China Agricultural University; WANG, WEI - China Agricultural University

Submitted to: Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/16/2023
Publication Date: 1/19/2023
Citation: Guo, X., Jia, B., Zhang, H., Ni, X., Zhuang, H., Lu, Y., Wang, W. 2023. Evaluation of Aspergillus flavus growth and detection of aflatoxin B1 content on maize agar culture medium using Vis/NIR hyperspectral imaging. Agriculture. 13. Article 237. https://doi.org/10.3390/agriculture13020237.
DOI: https://doi.org/10.3390/agriculture13020237

Interpretive Summary: The dynamic process of development and growth of fungus growth and subsequently aflatoxin production and accumulation in lab culture were examined using a visible/near-infrared hyper-spectral imaging system. First, the growth pattern of fungus and aflatoxin production were monitored in lab culture for five days with one-day interval. Second, the principal component analysis and independent component analysis were combined with a scatter map to analyze the fungal growth. The results showed that the scatter map method could distinguish the fungal growth time. Third, regression models to predict aflatoxin accumulation using hyper-spectral images were developed, where different pre-processing methods and key wavelengths were compared. The partial least squares regression model achieved the best regression performance for calibration and validation. Finally, a spatial map of aflatoxin concentration was created using the model established in this study. The spatial regularity of aflatoxin concentration was comparable to the measurement performed by high-performance liquid chromatography. The study results demonstrated the potential of utilizing the visible/near-infrared hyper-spectral imaging system to characterize fungal growth and aflatoxin production on culture media over time.

Technical Abstract: The dynamic growth process of Aspergillus flavus Link and the aflatoxin B1 (AFB1) accumulation in culture media were investigated with a visible/near-infrared hyperspectral imaging (Vis/NIR HSI) system in the range of 400 to 1000 nm. First, the growth of A. flavus and the synthesis pattern of AFB1 were monitored on maize agar medium (MAM) culture for 120 h with a 24-h time-lapse imaging interval. Second, the principal component analysis (PCA) and independent component analysis (ICA) were combined with an N-dimensional (ND) scatter map to analyze the A. flavus growth. The results showed that the ICA-ND scatter map method could distinguish the A. flavus growth time. Third, regression models to predict the AFB1 accumulation using hyperspectral images were developed by comparing different pre-processing methods and key wavelengths. The successive projection algorithm (SPA) was used to extract the key wavelengths. The experimental results indicated that the standard normal variate transformation (SNV) with the partial least squares regression (PLSR) achieved the best regression performance of R_c^2 value of 0.82-0.98 for calibration and R_v^2 value of 0.81-0.93 for validation. Finally, a spatial map of AFB1 concentration was created using the PLSR model. The spatial regularity of AFB1 concentration was comparable to the measurement performed using high-performance liquid chromatography (HPLC). The study results demonstrated the potential of the Vis/NIR HSI to characterize the growth of A. flavus growth and the concentration of AFB1 on culture media over time.