Machine learning for the meta-analyses of microbial pathogens’ volatile signatures.

Palma, S. I. C. J., Traguedo, A. P., Porteira, A. R., Frias, M. J., Gamboa, H., & Roque, A. C. A. (2018). Machine learning for the meta-analyses of microbial pathogens’ volatile signatures. Scientific Reports, 8, Article 3360. https://doi.org/10.1038/s41598-018-21647-6

This study leverages machine learning algorithms, particularly support vector machines and feature selection tools, to identify microbial volatile organic compounds (VOCs) that can differentiate between human pathogens. Data from studies published between 1977 and 2016 were analyzed, identifying 18 VOCs that can predict the presence of 11 microbial pathogens with 77% accuracy and precision between 62% and 100%. Each pathogen has a distinct set of VOCs, with a prediction accuracy of 86–90%. This approach improves non-invasive diagnostic tools, aiding in the development of artificial olfaction devices for infectious disease detection.