Authors
Yingxiao Peng, Xinyue Du, Hua Xia, Jiazheng Sun, Feng Gao, Yi Sun, Jiayi Wang, Xiangyun Ma, Yunpeng Yang, Qifeng Li
Published in
Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy. Volume 368. Issue Pt 1. Pages 129121. Nov 09, 2026. Epub Nov 09, 2026.
Abstract
Raman spectroscopy provides an effective method for the rapid detection of foodborne bacteria. However, single-cell Raman signals are weak, resulting in low classification accuracy. We present an Adaptive Frequency Mask Reconstruction (AFMR) method to improve bacterial classification accuracy in complex single-cell detection environments. The AFMR method employs an Adaptive Frequency Mask Reconstruction Network (AFMR-net) to learn and extract time-frequency features from Raman spectra spectrograms. The AFMR-net is trained using colony-level bacterial data to overcome the challenge posed by the scarcity of high-quality single-cell bacterial data. After the denoising process using the AFMR method, the recognition accuracy of the four Bacillus species increases from 77.7% to 95.8%. Our method improves the distinguishability of the four Bacillus species, establishing a solid foundation for the application of single-cell Raman spectroscopy in the fast, accurate detection of foodborne bacteria.
PMID:
43033756
Bibliographic data and abstract were imported from PubMed on 15 Nov 2026.
Read full publication at:
Please sign in
to see all details.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 6
- Comments 0