I’m still thinking about how to build a good database. I’d like to get data from proper labs who have access or can identify a pure sample, so there is a base data set for all substances to compare against as accessing pure samples illicitly is practically impossible.

From there, yes I would like to build data for common cutting agents & process impurities to determine sample purity.


In terms of what substances I’d be looking at: I’d start with MDMA and it’s numerous impersonators as well as the many things commonly found in “ecstasy”.

That’s a big task in itself, but the data would be applicable to a wider application of identifying almost any drug.


In theory, these should be fairly simple samples to analyse as, with mdma for example, it’s only one chemical, rather than a plant which may contain thousands of different chemicals. Right?

I’m only just starting to get my head around this device so there’s a long way to go before I actually start building a database & model.