Advanced machine learning for the characterisation of controlled substances using Raman spectroscopy

The aim of this project is to a develop a new class of instrument for real-time drug identification by combining

Raman technology with the latest machine learning software tools. This will enable new identification

databases to be developed, allowing small variations in drug composition to be identified and alerting the

authorities to counterfeit products and/or illegal and harmful drugs. This novel approach could enable the

development of new hand-held instrumentation for use in the homeland security, customs & excise,

pharmaceutical andr biological services sectors within 3 years. Within this study, the team will focus on the

production of a proof-of-concept system that will be tested against known samples. During the programme,

potential customers in the homeland security services sector will be consulted to ensure the development is

targeted at the key markets.aiting Public Project Summary