Electrochemical “Super-Fingerprinting” in Combination with Machine Learning for the On-Site Detection of Illicit Drugs

Electrochemical multidrug detection achieved high specificity (∼90%), sensitivity (∼93%), and accuracy (∼92%).

• The method combines data from four different pH conditions to improve detection.
• Machine learning techniques helped create a 'super-fingerprint' from electrochemical signals.
• Cocaine, heroin, ketamine, amphetamine, methamphetamine, and MDMA were successfully detected.
• The approach also identified 24 adulterants or cutting agents in samples.
• This technique may enhance on-site drug testing in complex real-world samples.

AI simplified