Image created by Dr. Michael J. Miller
Healthcare-associated infections remain a major global challenge, particularly as antimicrobial resistance continues to rise. Rapid and accurate detection of bacteria is essential, not only to treat patients effectively, but also to prevent the spread of infection.
INL researchers Susana Costa, Hedieh Mahmoodnia, Fábio Gonçalves, Adelaide Miranda and Pieter De Beule, in collaboration with INESC TEC, have developed a new approach that could transform how bacterial infections are identified. The research was published in Scientific Reports.
Instead of relying on traditional methods that can take days and require complex laboratory procedures, the team focused on something bacteria naturally produce: volatile organic compounds. These are small molecules released during bacterial metabolism, i.e. as bacteria live and grow, forming a unique chemical “fingerprint” for each species.
To capture these fingerprints, INL researchers developed a real-time sensing system based on a photoionization detector. By using multiple light sources, the system captures distinct signal patterns from each bacteria species.
Artificial intelligence then learns to recognise these patterns, allowing accurate identification of the bacteria.
Susana Costa explains that “by converting the signals into image-like representations, we trained a neural network to recognise and differentiate between bacterial species. This approach enables accurate recognition, while reducing the need for large training datasets, which are often difficult and time-consuming to obtain.”
The detection system was able to sense and distinguish clinically relevant bacteria, including Escherichia coli, Staphylococcus aureus, Pseudomonas aeruginosa, and Klebsiella pneumoniae, at low concentrations and in real time.
By combining advanced sensing with AI-driven analysis, this work opens new possibilities for faster, simpler, and more adaptable diagnostic tools. In a near future, such technologies could support earlier detection of infections, more targeted treatments, and improved patient outcomes.
Reference
Costa, S.P., Cardoso, A., Mahmoodnia, H. et al. Bacterial species differentiation via real-time detection of microbial volatile organic compounds using a wavelength multiplexed photoionization detector and AI image-based analysis. Sci Rep 16, 15924 (2026). https://doi.org/10.1038/s41598-026-46818-x
Abstract
Healthcare-associated infections (HCAIs) contribute significantly to global mortality, driven by the increasing antimicrobial resistance. Rapid, high-throughput bacterial detection is crucial for infection control and patient care. We report a real-time, multiplex lamp-based Photoionization Detector (PID) assisted by AI-image-based analysis for bacterial identification. Using four lamps with varying ionization energies, the sensor selectively ionizes VOCs emitted by bacteria, producing four distinct current curves for each target species (Escherichia coli, Staphylococcus aureus, Pseudomonas aeruginosa, and Klebsiella pneumoniae). These curves were transformed into image representations, capturing their spectral patterns for bacterial differentiation. A pre-trained ResNet-18 Convolutional Neural Network (CNN) within a Few-Shot Learning (FSL) framework extracted key features, enabling accurate (> 88%) bacterial differentiation even with limited labeled data. This sensor detected bacterial concentrations as low as 10² CFU and distinguished contamination levels. The synergistic integration of PID sensing with AI-driven analysis offers a powerful approach to rapid bacterial diagnostics, demonstrating strong potential for clinical implementation and improved patient care. This study marks an early step toward AI-based VOC sensing, where FSL acts as a proof-of-concept under data scarcity.