Classification of radix angelicae dahuricae from different regions by terahertz spectroscopy

In this work, terahertz time-domain spectroscopy was employed to establish the fingerprints of Radix Angelicae Dahuricae (RAD). Principal component analysis (PCA) and k-subspace clustering were employed to study the similarities and dissimilarities among the 120 RAD samples. The methods can be successfully used to classify these samples according to their regions. The results show that the referred technologies can be efficiently used to classify RAD into groups, which provides a method for pattern recognition of other traditional Chinese medicine.