Logo image
Deep Learning Terahertz Spectroscopy for Non-Invasive Traditional Chinese Medicine Identification
會議論文

Deep Learning Terahertz Spectroscopy for Non-Invasive Traditional Chinese Medicine Identification

Jui-Chi Lin, Chia-Ming Mai, Shao-Shuan Wu, Wen-Tai Li 和 Shang-Hua Yang
International Conference on Infrared, Millimeter, and Terahertz Waves (Print), 頁碼.1-2
IEEE
2024 49th International Conference on Infrared, Millimeter, and Terahertz Waves (IRMMW-THz) (Perth, Australia, 01/09/2024–06/09/2024)
01/09/2024
Web of Science ID: WOS:001334520200343

摘要

artificial neural network Artificial neural networks Databases Deep learning ginseng Monitoring Protocols Quality control Reproducibility of results terahertz time-domain spectroscopy Time-domain analysis Training Spectroscopy
In traditional Chinese medicine (TCM), quality control involves accurately distinguishing between species, notably the widely utilized herbs - Panax ginseng and Panax quinquefolius. Terahertz time-domain spectroscopy (THz-TDS) emerges as a non-destructive and time-efficient tool, promising application not only in academic research but also in practical settings. Despite of its potential, an integrated system for standardized protocols, spectral databases, and data analysis is lacking. In this research, we focus on measurement quality and reproducibility by developing a protocol. Next by training an artificial neural network (ANN), differentiation of the two Panax species have been achieved with 92.86% accuracy using 6-fold cross-validation. Combine THz-TDS measurement with deep learning for prediction, we advanced quality monitoring and species differentiation in TCM.

相關連結

指標

1 檢視次數

詳細資料

Logo image