Research at the University of Lausanne’s School of Criminal Sciences combines infrared spectroscopy and AI for fast, reliable detection in the field.
The research, led by Hervé Rais, Olivier Delémont and Pierre Esseiva, involves deploying portable and rapid techniques for the detection and analysis of falsified medicines. Thanks to portable technology based on near-infrared spectral analysis, promoted by NIRLAB (Nirlab.com), a start-up from the University of Lausanne, it is possible to decentralize analysis techniques in order to provide field workers with real-time results.
An emblematic case: Viagra
The study focused in particular on Viagra, which accounts for the vast majority of cases of drug falsification in Switzerland. Often purchased from online pharmacies, these products are not approved by the supervisory authorities. Their exact composition and the dosage of the active substance are therefore not certified.
The aim was to find a way of determining, in real time and non-destructively, whether Viagra specialties were genuine or adulterated, and in the latter case, to check whether the active substance was indeed in the advertised dosages.
Results and scientific collaboration
To train the analysis algorithms, it is essential to have the exact composition of the tablets. This is the background to a long-standing collaboration with Professor Serge Rudaz and his team at the University of Geneva’s School of Pharmacy. of Geneva, who characterized the samples used to develop the statistical models for this study.
Given the high accuracy of the models developed, the results obtained using portable technology based on near-infrared (NIR) spectroscopy are comparable to those of reference laboratory analytical methods. The study demonstrated that machine learning models applied to spectral data can not onlyidentify falsified tablets with 100% accuracy, but also accurately estimate the quantity of active molecule.
Deployment beyond Viagra
This approach is also being deployed on a wider scale for other specialties, such as antimalarials or analgesics, throughthis approach is also being deployed on a wider scale for other specialties, such as anti-malarial drugs or painkillers, through collaborations with various leading institutions involved in the fight against this phenomenon, or active in the field in African countries. The ultimate aim is to offer those working in the field the opportunity to test the products that are or will be made available to the population. Being able to carry out analyses in real time, quickly, reliably, non-destructively and without the need for specific training, is a major step forward in the drive to bring greater safety to the consumption of pharmaceutical products.
Reference
Hervé Rais, Pierre Esseiva, Olivier Delémont, Cédric Schelling, Stefan Stanojevic, Serge Rudaz, Florentin Coppey, Rapid detection and quantification of falsified Viagra using cloud-based portable NIR technology and machine learning, Journal of Pharmaceutical and Biomedical Analysis, Volume 263, 2025
10.1016/j.jpba.2025.116940
