ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Culture-Independent Raman Spectroscopic Identification of Bacterial Pathogens from Clinical Samples Using Deep Transfer Learning

Singh, S and Kumbhar, D and Reghu, D and Venugopal, SJ and Rekha, PT and Mohandas, S and Rao, S and Rangaiah, A and Chunchanur, SK and Saini, DK and Umapathy, S (2022) Culture-Independent Raman Spectroscopic Identification of Bacterial Pathogens from Clinical Samples Using Deep Transfer Learning. In: Analytical Chemistry, 94 (42). pp. 14745-14754.

[img] PDF
ana_che_ 94-42_14745 - 14754_2022.pdf - Published Version
Restricted to Registered users only

Download (5MB) | Request a copy
Official URL: https://doi.org/10.1021/acs.analchem.2c03391

Abstract

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. © 2022 American Chemical Society. All rights reserved.

Item Type: Journal Article
Publication: Analytical Chemistry
Publisher: American Chemical Society
Additional Information: The copyright for this article belongs to American Chemical Society
Keywords: Bacteria; Classification (of information); Deep learning; Nucleic acids; Pathogens, Bacterial identifications; Bacterial pathogens; Clinical isolates; Clinical samples; Data set; Raman spectroscopic; Rapid identification; Spectra's; Spectroscopic identification; Transfer learning, Raman spectroscopy
Department/Centre: Division of Biological Sciences > Microbiology & Cell Biology
Division of Biological Sciences > Molecular Reproduction, Development & Genetics
Division of Biological Sciences > Centre for Infectious Disease Research
Division of Chemical Sciences > Inorganic & Physical Chemistry
Division of Interdisciplinary Sciences > Centre for Biosystems Science and Engineering
Division of Physical & Mathematical Sciences > Instrumentation Appiled Physics
Date Deposited: 13 Dec 2022 05:49
Last Modified: 13 Dec 2022 05:49
URI: https://eprints.iisc.ac.in/id/eprint/78467

Actions (login required)

View Item View Item