Authenticating botanical materials can be challenging, especially when large numbers of samples must be evaluated or differences between chromatographic fingerprints are subtle.

Our latest application note demonstrates how One-Class Support Vector Machine (OC-SVM) classification in visionCATS can be combined with HPTLC fingerprinting to support a standardized, data-driven approach to authenticity assessment, using red Asian ginseng as an example.

Technical summary:

• Uses authenticated samples to establish the reference class
• Does not require examples of every possible adulterant or substitute
• Evaluates complete HPTLC fingerprint profiles
• Flags samples that do not conform to the authenticated reference model