RESUMEN

The rapid identification of bacterial pathogens in clinical samples like blood, urine, pus, and sputum is the need of the hour. Conventional bacterial identification methods like culturing and nucleic acid-based amplification have limitations like poor sensitivity, high cost, slow turnaround time, etc. Raman spectroscopy, a label-free and noninvasive technique, has overcome these drawbacks by providing rapid biochemical signatures from a single bacterium. Raman spectroscopy combined with chemometric methods has been used effectively to identify pathogens. However, a robust approach is needed to utilize Raman features for accurate classification while dealing with complex data sets such as spectra obtained from clinical isolates, showing high sample-to-sample heterogeneity. In this study, we have used Raman spectroscopy-based identification of pathogens from clinical isolates using a deep transfer learning approach at the single-cell level resolution. We have used the data-augmentation method to increase the volume of spectra needed for deep-learning analysis. Our ResNet model could specifically extract the spectral features of eight different pathogenic bacterial species with a 99.99% classification accuracy. The robustness of our model was validated on a set of blinded data sets, a mix of cultured and noncultured bacterial isolates of various origins and types. Our proposed ResNet model efficiently identified the pathogens from the blinded data set with high accuracy, providing a robust and rapid bacterial identification platform for clinical microbiology.

Asunto(s)

Ácidos Nucleicos , Espectrometría Raman , Espectrometría Raman/métodos , Bacterias , Aprendizaje Automático , Extractos Vegetales

RESUMEN

Background: Ongoing need of alternative strategies for SARS-CoV-2 detection is undeniable. Self-collected samples without viral transport media (VTM), coupled with simple nucleic acid extraction methods for SARS-CoV-2 PCR are beneficial. Objectives: To evaluate results of SARS-CoV-2 PCR using simple nucleic acid extraction methods from self -collected saliva and oral swabs without VTM. Methods: A cross-sectional single-centre study was conducted on 125 participants (101 SARS-CoV-2 positive cases and 24 controls). PCR was performed following five simple nucleic acid extraction methods on self -collect saliva and oral swabs without VTM and results were compared with gold standard PCR. For saliva, kit-based extraction (SKE), Proteinase K and Heat extraction (SPHE), only Heat extraction (SHE) methods and for dry oral swabs, Proteinase K and Heat extraction (DPHE) and only Heat extraction (DHE) was performed. Results: SARS-CoV-2 was detected in self-collected saliva and oral swabs. 93.07% were correctly classified as positive by SKE, 69.31% by SHE, 67.33% by SPHE, 67.33% by DPHE and 55.45% by DHE. Discriminant power of SKE was significantly higher than other methods (p-value < 0.001) with good- fair agreement of alternate extraction methods against gold standard. Conclusion: Combination of self-collected saliva/ oral-swab without VTM and alternative RNA extraction methods offer a simplified, economical substitute strategy for SARS-CoV-2 detection.