How to achieve auto-identification in Raman analysis by spectral feature extraction &amp; Adaptive Hypergraph

Xie, Yi; You, Qiaobei; Dai, Pingyang; Wang, Shuyi; Hong, Peiyi; Liu, Guokun; Yu, Jun; Sun, Xiaorui; Zeng, Yongming

doi:10.1016/j.saa.2019.04.078

Cited by 8 publications

(3 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, for phototrophs such as microalgae, potential remains elusive . Specifically, for photosynthetic cells, two types of SCRS including “pigment spectrum” (PS; acquired before quenching) and “whole spectrum” (WS; collected after quenching) are possible, yet past studies have generally sampled either PS or WS, but not both; − thus, the optimal strategy to extract information from SCRS remains to be developed. ,− Moreover, the breadth of sampling over the microalgal space has been extremely limited, as only a few species were included in tests for phylogenetic classification , or metabolite profiling. − Furthermore, SCRS are sensitive to not only “phylogeny” but also the “state” of a cell, yet existing experimental designs have generally failed to distinguish them. Consequentially, how broadly applicable or reliable the approach is for microalgae remains unclear.…”

Section: Introductionmentioning

confidence: 99%

Culture-Free Identification and Metabolic Profiling of Microalgal Single Cells via Ensemble Learning of Ramanomes

Baladehi

Hekmatara

et al. 2021

Anal. Chem.

View full text Add to dashboard Cite

Microalgae are among the most genetically and metabolically diverse organisms on earth, yet their identification and metabolic profiling have generally been slow and tedious. Here, we established a reference ramanome database consisting of single-cell Raman spectra (SCRS) from >9000 cells of 27 phylogenetically diverse microalgal species, each under stationary and exponential states. When combined, prequenching (“pigment spectrum” (PS)) and postquenching (“whole spectrum” (WS)) signals can classify species and states with 97% accuracy via ensemble machine learning. Moreover, the biosynthetic profile of Raman-sensitive metabolites was unveiled at single cells, and their interconversion was detected via intra-ramanome correlation analysis. Furthermore, not-yet-cultured cells from the environment were functionally characterized via PS and WS and then phylogenetically identified by Raman-activated sorting and sequencing. This PS–WS combined approach for rapidly identifying and metabolically profiling single cells, either cultured or uncultured, greatly accelerates the mining of microalgae and their products.

show abstract

Section: Introductionmentioning

confidence: 99%

Culture-Free Identification and Metabolic Profiling of Microalgal Single Cells via Ensemble Learning of Ramanomes

Baladehi

Hekmatara

et al. 2021

Anal. Chem.

View full text Add to dashboard Cite

show abstract

“…Raman spectra of cefuroxime axetil tablets and the mixture of pyrene and cyclodextrin (2 mM), as well as the Raman mapping of cefuroxime axetil and cervical cancer cells, were obtained using the Raman-11 Nanophoton system with a 785 nm excitation laser at 500 mW . The SERS spectra of Acid Red 18 in tea, chips, sweets, and yogurt were measured using a B&W Tek i-Raman spectrometer with a 785 nm excitation laser at 500 mW …”

Section: Methodsmentioning

confidence: 99%

“…24 The SERS spectra of Acid Red 18 in tea, chips, sweets, and yogurt were measured using a B&W Tek i-Raman spectrometer with a 785 nm excitation laser at 500 mW. 33 Methodology. Figure 1a shows the flow of the PEER algorithm with four steps.…”

Section: ■ Experiments and Methodologymentioning

confidence: 99%

Developing a Peak Extraction and Retention (PEER) Algorithm for Improving the Temporal Resolution of Raman Spectroscopy

et al. 2021

Self Cite

View full text Add to dashboard Cite

In spectroscopic analysis, push-to-the-limit sensitivity is one of the important topics, particularly when facing the qualitative and quantitative analyses of the trace target. Normally, the effective recognition and extraction of weak signals are the first key steps, for which there has been considerable effort in developing various denoising algorithms for decades. Nevertheless, the lower the signal-to-noise ratio (SNR), the greater the deviation of the peak height and shape during the denoising process. Therefore, we propose a denoising algorithm along with peak extraction and retention (PEER). First, both the first and second derivatives of the Raman spectrum are used to determine Raman peaks with a high SNR whose peak information is kept away from the denoising process. Second, an optimized window smoothing algorithm is applied to the left part of the Raman spectrum, which is combined with the untreated Raman peaks to obtain the denoised Raman spectrum. The PEER algorithm is demonstrated with much better signal extraction and retention and successfully improves the temporal resolution of Raman imaging of a living cell by at least 1 order of magnitude higher than those by traditional algorithms.

show abstract

Recent Progresses in Machine Learning Assisted Raman Spectroscopy

Jiang

et al. 2023

Advanced Optical Materials

View full text Add to dashboard Cite

With the development of Raman spectroscopy and the expansion of its application domains, conventional methods for spectral data analysis have manifested many limitations. Exploring new approaches to facilitate Raman spectroscopy and analysis has become an area of intensifying focus for research. It has been demonstrated that machine learning techniques can more efficiently extract valuable information from spectral data, creating unprecedented opportunities for analytical science. This paper outlines traditional and more recently developed statistical methods that are commonly used in machine learning (ML) and ML‐algorithms for different Raman spectroscopy‐based classification and recognition applications. The methods include Principal Component Analysis, K‐Nearest Neighbor, Random Forest, and Support Vector Machine, as well as neural network‐based deep learning algorithms such as Artificial Neural Networks, Convolutional Neural Networks, etc. The bulk of the review is dedicated to the research advances in machine learning applied to Raman spectroscopy from several fields, including material science, biomedical applications, food science, and others, which reached impressive levels of analytical accuracy. The combination of Raman spectroscopy and machine learning offers unprecedented opportunities to achieve high throughput and fast identification in many of these application fields. The limitations of current studies are also discussed and perspectives on future research are provided.

show abstract

How to achieve auto-identification in Raman analysis by spectral feature extraction & Adaptive Hypergraph

Cited by 8 publications

References 19 publications

Culture-Free Identification and Metabolic Profiling of Microalgal Single Cells via Ensemble Learning of Ramanomes

Culture-Free Identification and Metabolic Profiling of Microalgal Single Cells via Ensemble Learning of Ramanomes

Developing a Peak Extraction and Retention (PEER) Algorithm for Improving the Temporal Resolution of Raman Spectroscopy

Recent Progresses in Machine Learning Assisted Raman Spectroscopy

Contact Info

Product

Resources

About