Recovery of Raman spectra with low signal-to-noise ratio using Wiener estimation
Raman spectroscopy is a powerful non-destructive technique for qualitatively and quantitatively characterizing materials. However, noise often obscures interesting Raman peaks due to the inherently weak Raman signal, especially in biological samples. In this study, we develop a method based on spect...
| Main Authors: | , , , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2014
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/80910 http://hdl.handle.net/10220/19743 |
| _version_ | 1826115292659449856 |
|---|---|
| author | Chen, Shuo Lin, Xiaoqian Yuen, Clement Padmanabhan, Saraswathi Beuerman, Roger W. Liu, Quan |
| author2 | School of Chemical and Biomedical Engineering |
| author_facet | School of Chemical and Biomedical Engineering Chen, Shuo Lin, Xiaoqian Yuen, Clement Padmanabhan, Saraswathi Beuerman, Roger W. Liu, Quan |
| author_sort | Chen, Shuo |
| collection | NTU |
| description | Raman spectroscopy is a powerful non-destructive technique for qualitatively and quantitatively characterizing materials. However, noise often obscures interesting Raman peaks due to the inherently weak Raman signal, especially in biological samples. In this study, we develop a method based on spectral reconstruction to recover Raman spectra with low signal-to-noise ratio (SNR). The synthesis of narrow-band measurements from low-SNR Raman spectra eliminates the effect of noise by integrating the Raman signal along the wavenumber dimension, which is followed by spectral reconstruction based on Wiener estimation to recover the Raman spectrum with high spectral resolution. Non-negative principal components based filters are used in the synthesis to ensure that most variance contained in the original Raman measurements are retained. A total of 25 agar phantoms and 20 bacteria samples were measured and data were used to validate our method. Four commonly used de-noising methods in Raman spectroscopy, i.e. Savitzky-Golay (SG) algorithm, finite impulse response (FIR) filtration, wavelet transform and factor analysis, were also evaluated on the same set of data in addition to the proposed method for comparison. The proposed method showed the superior accuracy in the recovery of Raman spectra from measurements with extremely low SNR, compared with the four commonly used de-noising methods. |
| first_indexed | 2024-10-01T03:53:06Z |
| format | Journal Article |
| id | ntu-10356/80910 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T03:53:06Z |
| publishDate | 2014 |
| record_format | dspace |
| spelling | ntu-10356/809102023-12-29T06:47:07Z Recovery of Raman spectra with low signal-to-noise ratio using Wiener estimation Chen, Shuo Lin, Xiaoqian Yuen, Clement Padmanabhan, Saraswathi Beuerman, Roger W. Liu, Quan School of Chemical and Biomedical Engineering DRNTU::Engineering Raman spectroscopy is a powerful non-destructive technique for qualitatively and quantitatively characterizing materials. However, noise often obscures interesting Raman peaks due to the inherently weak Raman signal, especially in biological samples. In this study, we develop a method based on spectral reconstruction to recover Raman spectra with low signal-to-noise ratio (SNR). The synthesis of narrow-band measurements from low-SNR Raman spectra eliminates the effect of noise by integrating the Raman signal along the wavenumber dimension, which is followed by spectral reconstruction based on Wiener estimation to recover the Raman spectrum with high spectral resolution. Non-negative principal components based filters are used in the synthesis to ensure that most variance contained in the original Raman measurements are retained. A total of 25 agar phantoms and 20 bacteria samples were measured and data were used to validate our method. Four commonly used de-noising methods in Raman spectroscopy, i.e. Savitzky-Golay (SG) algorithm, finite impulse response (FIR) filtration, wavelet transform and factor analysis, were also evaluated on the same set of data in addition to the proposed method for comparison. The proposed method showed the superior accuracy in the recovery of Raman spectra from measurements with extremely low SNR, compared with the four commonly used de-noising methods. ASTAR (Agency for Sci., Tech. and Research, S’pore) Accepted Version 2014-06-13T03:42:25Z 2019-12-06T14:17:12Z 2014-06-13T03:42:25Z 2019-12-06T14:17:12Z 2014 2014 Journal Article Chen, S., Lin, X., Yuen, C., Padmanabhan, S., Beuerman, R. W., & Liu, Q. (2014). Recovery of Raman spectra with low signal-to-noise ratio using Wiener estimation. Optics Express, 22(10), 12102-12114 . 1094-4087 https://hdl.handle.net/10356/80910 http://hdl.handle.net/10220/19743 10.1364/OE.22.012102 en Optics express © 2014 Optical Society of America. This is the author created version of a work that has been peer reviewed and accepted for publication by Optics express, Optical Society of America. It incorporates referee’s comments but changes resulting from the publishing process, such as copyediting, structural formatting, may not be reflected in this document. The published version is available at: [http://dx.doi.org/10.1364/OE.22.012102]. application/pdf |
| spellingShingle | DRNTU::Engineering Chen, Shuo Lin, Xiaoqian Yuen, Clement Padmanabhan, Saraswathi Beuerman, Roger W. Liu, Quan Recovery of Raman spectra with low signal-to-noise ratio using Wiener estimation |
| title | Recovery of Raman spectra with low signal-to-noise ratio using Wiener estimation |
| title_full | Recovery of Raman spectra with low signal-to-noise ratio using Wiener estimation |
| title_fullStr | Recovery of Raman spectra with low signal-to-noise ratio using Wiener estimation |
| title_full_unstemmed | Recovery of Raman spectra with low signal-to-noise ratio using Wiener estimation |
| title_short | Recovery of Raman spectra with low signal-to-noise ratio using Wiener estimation |
| title_sort | recovery of raman spectra with low signal to noise ratio using wiener estimation |
| topic | DRNTU::Engineering |
| url | https://hdl.handle.net/10356/80910 http://hdl.handle.net/10220/19743 |
| work_keys_str_mv | AT chenshuo recoveryoframanspectrawithlowsignaltonoiseratiousingwienerestimation AT linxiaoqian recoveryoframanspectrawithlowsignaltonoiseratiousingwienerestimation AT yuenclement recoveryoframanspectrawithlowsignaltonoiseratiousingwienerestimation AT padmanabhansaraswathi recoveryoframanspectrawithlowsignaltonoiseratiousingwienerestimation AT beuermanrogerw recoveryoframanspectrawithlowsignaltonoiseratiousingwienerestimation AT liuquan recoveryoframanspectrawithlowsignaltonoiseratiousingwienerestimation |