Revolutionizing Gm Rice Seed Detection: Novel Spectral Phenotyping And Deep Learning Approach Unveiled

This research article published in Plant Phenomics in July 2023 examines the use of advanced spectroscopic techniques, including near infrared and terahertz, to identify genetically modified (GM) organisms. Machine learning algorithms, particularly deep learning, have improved the accuracy of these methods. The study found that the CascadeSeed-1 model, using NIR and terahertz spectra, was more accurate in variety discrimination than other machine learning models. The CascadeSeed-2 model was also found to have high accuracy in GM status identification, with terahertz spectra-based models outperforming NIR-based ones.