Abstract
The rapid and accurate identification of pathogenic bacteria is crucial for combating the growing threat of antibiotic resistance, nosocomial infections, and food safety concerns. This study presents a novel and comprehensive comparison of two vibrational spectroscopic techniques - attenuated total reflection Fourier transform infrared (ATR-FTIR) spectroscopy and a low-cost miniature near-infrared (NIR) spectrometer - for distinguishing Gram-positive and Gram-negative bacterial samples grown using the same stock media solution. This is the first report of NIR spectroscopy being applied to differentiate Gram-positive and Gram-negative bacteria, as well as the first direct comparison of ATR-FTIR and NIR for the combined multimodal analysis of clinical bacterial isolates. Using a data set of five Gram-positive and seven Gram-negative species and recording spectra in triplicate, the study employed advanced data fusion and multivariate analysis techniques to classify the spectra and facilitate NIR band assignment. 2D correlation analysis revealed strong positive correlations between key spectral markers identified in both modalities. Partial least-squares- and support vector machine discriminant analysis models were validated using a methodology based on 100 repeated random sampling of calibration and test sets. Models demonstrated that both the standalone ATR-FTIR and the combined ATR-FTIR/NIR approach achieved exceptional classification accuracy (>98%) in differentiating the two bacterial groups. Differences observed in the spectra were attributed to the distinct cell wall compositions of Gram-Positive and Gram-negative bacteria. Notably, the low-cost NIR technique also showed promising performance, with classification accuracy values above 90%. The findings highlight the potential of these rapid, noninvasive, and cost-effective vibrational spectroscopic techniques, particularly the NIR method, for point-of-care applications in clinical microbiology and food safety monitoring. The combination of ATR-FTIR and NIR data further enhances the robustness and reliability of bacterial identification, paving the way for broader adoption of these advanced analytical tools in various healthcare and food safety settings.
Original language | English |
---|---|
Pages (from-to) | 18392-18400 |
Number of pages | 9 |
Journal | Analytical Chemistry |
Volume | 96 |
Issue number | 46 |
DOIs | |
Publication status | Published - 19 Nov 2024 |
Funding
This work is supported by an Australian Research Council (ARC) Future Fellowship grant FT120100926 and an ARC Discovery Project grant DP180103484. The authors acknowledge Finlay Shanks for instrumental support. Thulya\u2019s research was supported by the Monash University- University of Bath global PhD programme. Furthermore, Thulya would like to acknowledge the Australian Institute of Nuclear Science and Engineering (AINSE)\u2019s postgraduate research award (PGRA) 2021. D.P.-G. acknowledges the financial support of the 2019 Ramo\u0301n y Cajal (RYC) Contract Aids (RYC2019- 026556-I) and Grant RPID2020-119326RA-I0 funded by MCIN/AEI /10.13039/501100011033 and FSE \u201CEl FSE invierte en tu futuro\u201D.
Funders | Funder number |
---|---|
Australian Research Council | FT120100926, DP180103484 |
Monash University- University of Bath | RPID2020-119326RA-I0, RYC2019- 026556-I |
Ministerio de Ciencia e Innovación | |
Agencia Estatal de Investigación | |
Faculty of Science and Engineering, University of Manchester | |
Australian Institute of Nuclear Science and Engineering | |
Finlay Shanks |
Keywords
- Amides
- Bacteria
- Infrared light
- infrared spectroscopy
- Mathematical methods
ASJC Scopus subject areas
- Analytical Chemistry