How to pre-process Raman spectra for reliable and stable models?

Fusing 1H NMR and Raman experimental data for the improvement of wine recognition models

The present study proposes the development of new wine recognition models based on Artificial Intelligence (AI) applied to the mid-level data fusion of 1H NMR and Raman data. In this regard, a supervised machine learning method, namely Support Vector Machines (SVMs), was applied for classifying wine samples with respect to the cultivar, vintage, and geographical origin. Because the association between the two data sources generated an input space with a high dimensionality, a feature selection algorithm was employed to identify the most relevant discriminant markers for each wine classification criterion, before SVM modeling. The proposed data processing strategy allowed the classification of the wine sample set with accuracies up to 100% in both cross-validation and on an independent test set and highlighted the efficiency of 1H NMR and Raman data fusion as opposed to the use of a single-source data for differentiating wine concerning the cultivar and vintage.

Spectral Zones-Based SHAP/LIME: Enhancing Interpretability in Spectral Deep Learning Models Through Grouped Feature Analysis

Interpretability is just as important as accuracy when it comes to complex models, especially in the context of deep learning models. Explainable artificial intelligence (XAI) approaches have been developed to address this problem. The literature on XAI for spectroscopy mainly emphasizes independent feature analysis with limited application of zone analysis. Individual feature analysis methods, such as shapley additive explanations (SHAP) and local interpretable model-agnostic explanations (LIME), have limitations due to their dependence on perturbations. These methods measure how AI models respond to sudden changes in the individual feature values. While they can help identify the most impactful features, the abrupt shifts introduced by replacing these values with zero or the expected ones may not accurately represent real-world scenarios. This can lead to mathematical and computational interpretations that are neither physically realistic nor intuitive to humans. Our proposed method does not rely on individual disturbances. Instead, it targets "spectral zones" to directly estimate the effect of group disturbances on a trained model. Consequently, factors such as sample size, hyperparameter selection, and other training-related considerations are not the primary focus of the XAI methods. To achieve this, we have developed a modified version of LIME and SHAP capable of performing group perturbations, enhancing explainability and realism while minimizing noise in the plots used for interpretability. Additionally, we employed an efficient approach to calculate spectral zones for complex spectra with indistinct spectral boundaries. Users can also define the zones themselves using their domain-specific knowledge.

Standardized Electric-Field-Resolved Molecular Fingerprinting

Field-resolved infrared spectroscopy (FRS) of impulsively excited molecular vibrations can surpass the sensitivity of conventional time-integrating spectroscopies, owing to a temporal separation of the molecular signal from the noisy excitation. However, the resonant response carrying the molecular signal of interest depends on both the amplitude and phase of the excitation, which can vary over time and across different instruments. To date, this has compromised the accuracy with which FRS measurements could be compared, which is a crucial factor for practical applications. Here, we utilize a data processing procedure that overcomes this shortcoming while preserving the sensitivity of FRS. We validate the approach for aqueous solutions of molecules. The employed approach is compatible with established processing and evaluation methods for the analysis of infrared spectra and can be applied to existing spectra from databases, facilitating the spread of FRS to new molecular analytical applications.

SERS microscopy as a tool for comprehensive biochemical characterization in complex samples

Surface enhanced Raman scattering (SERS) spectra of biomaterials such as cells or tissues can be used to obtain biochemical information from nanoscopic volumes in these heterogeneous samples. This tutorial review discusses the factors that determine the outcome of a SERS experiment in complex bioorganic samples. They are related to the SERS process itself, the possibility to selectively probe certain regions or constituents of a sample, and the retrieval of the vibrational information in order to identify molecules and their interaction. After introducing basic aspects of SERS experiments in the context of biocompatible environments, spectroscopy in typical microscopic settings is exemplified, including the possibilities to combine SERS with other linear and non-linear microscopic tools, and to exploit approaches that improve lateral and temporal resolution. In particular the great variation of data in a SERS experiment calls for robust data analysis tools. Approaches will be introduced that have been originally developed in the field of bioinformatics for the application to omics data and that show specific potential in the analysis of SERS data. They include the use of simulated data and machine learning tools that can yield chemical information beyond achieving spectral classification.

Iterative Regression of Corrective Baselines (IRCB): A New Model for Quantitative Spectroscopy

In this work, a new model with broad utility for quantitative spectroscopy development is reported. A primary objective of this work is to create a novel modeling procedure that may allow for higher automation of the model development process. The fundamental concept is simple yet powerful even for complex spectra and is employed with no additional preprocessing. This approach is applicable for several types of spectroscopic data to develop regression models that have similar or greater quality than the current methods. The key modeling steps are a matrix transformation and subsequent feature selection process that are collectively referred to as iterative regression of corrective baselines (IRCB). The transformed matrix (Xtransform) is a linearized form of the original X data set. Features from Xtransform that are predictive of Y can be ranked and selected by ordinary least-squares regression. The best features (rows of Xtransform) are linear depictions of Y that can be utilized to develop regression models with several machine learning models. The IRCB workflow is first detailed by using a case study of Fourier transform infrared (FTIR) spectroscopy for prepared solutions of a three-component mixture. Next, IRCB is applied and compared to benchmark results for the 2006 "Chimiométrie" near-infrared spectroscopy (NIR) soil composition challenge and Raman measurements of a simulated nuclear waste slurry.

Improving skin cancer detection by Raman spectroscopy using convolutional neural networks and data augmentation

Background

Our previous studies have demonstrated that Raman spectroscopy could be used for skin cancer detection with good sensitivity and specificity. The objective of this study is to determine if skin cancer detection can be further improved by combining deep neural networks and Raman spectroscopy.

Patients and methods

Raman spectra of 731 skin lesions were included in this study, containing 340 cancerous and precancerous lesions (melanoma, basal cell carcinoma, squamous cell carcinoma and actinic keratosis) and 391 benign lesions (melanocytic nevus and seborrheic keratosis). One-dimensional convolutional neural networks (1D-CNN) were developed for Raman spectral classification. The stratified samples were divided randomly into training (70%), validation (10%) and test set (20%), and were repeated 56 times using parallel computing. Different data augmentation strategies were implemented for the training dataset, including added random noise, spectral shift, spectral combination and artificially synthesized Raman spectra using one-dimensional generative adversarial networks (1D-GAN). The area under the receiver operating characteristic curve (ROC AUC) was used as a measure of the diagnostic performance. Conventional machine learning approaches, including partial least squares for discriminant analysis (PLS-DA), principal component and linear discriminant analysis (PC-LDA), support vector machine (SVM), and logistic regression (LR) were evaluated for comparison with the same data splitting scheme as the 1D-CNN.

Results

The ROC AUC of the test dataset based on the original training spectra were 0.886±0.022 (1D-CNN), 0.870±0.028 (PLS-DA), 0.875±0.033 (PC-LDA), 0.864±0.027 (SVM), and 0.525±0.045 (LR), which were improved to 0.909±0.021 (1D-CNN), 0.899±0.022 (PLS-DA), 0.895±0.022 (PC-LDA), 0.901±0.020 (SVM), and 0.897±0.021 (LR) respectively after augmentation of the training dataset (p<0.0001, Wilcoxon test). Paired analyses of 1D-CNN with conventional machine learning approaches showed that 1D-CNN had a 1-3% improvement (p<0.001, Wilcoxon test).

Conclusions

Data augmentation not only improved the performance of both deep neural networks and conventional machine learning techniques by 2-4%, but also improved the performance of the models on spectra with higher noise or spectral shifting. Convolutional neural networks slightly outperformed conventional machine learning approaches for skin cancer detection by Raman spectroscopy.

A Study of High-Frequency Noise for Microplastics Classification Using Raman Spectroscopy and Machine Learning

Given the growing urge for plastic management and regulation in the world, recent studies have investigated the problem of plastic material identification for correct classification and disposal. Recent works have shown the potential of machine learning techniques for successful microplastics classification using Raman signals. Classification techniques from the machine learning area allow the identification of the type of microplastic from optical signals based on Raman spectroscopy. In this paper, we investigate the impact of high-frequency noise on the performance of related classification tasks. It is well-known that classification based on Raman is highly dependent on peak visibility, but it is also known that signal smoothing is a common step in the pre-processing of the measured signals. This raises a potential trade-off between high-frequency noise and peak preservation that depends on user-defined parameters. The results obtained in this work suggest that a linear discriminant analysis model cannot generalize properly in the presence of noisy signals, whereas an error-correcting output codes model is better suited to account for inherent noise. Moreover, principal components analysis (PCA) can become a must-do step for robust classification models, given its simplicity and natural smoothing capabilities. Our study on the high-frequency noise, the possible trade-off between pre-processing the high-frequency noise and the peak visibility, and the use of PCA as a noise reduction technique in addition to its dimensionality reduction functionality are the fundamental aspects of this work.

Raman-based PAT for VLP precipitation: systematic data diversification and preprocessing pipeline identification

Virus-like particles (VLPs) are a promising class of biopharmaceuticals for vaccines and targeted delivery. Starting from clarified lysate, VLPs are typically captured by selective precipitation. While VLP precipitation is induced by step-wise or continuous precipitant addition, current monitoring approaches do not support the direct product quantification, and analytical methods usually require various, time-consuming processing and sample preparation steps. Here, the application of Raman spectroscopy combined with chemometric methods may allow the simultaneous quantification of the precipitated VLPs and precipitant owing to its demonstrated advantages in analyzing crude, complex mixtures. In this study, we present a Raman spectroscopy-based Process Analytical Technology (PAT) tool developed on batch and fed-batch precipitation experiments of Hepatitis B core Antigen VLPs. We conducted small-scale precipitation experiments providing a diversified data set with varying precipitation dynamics and backgrounds induced by initial dilution or spiking of clarified Escherichia coli-derived lysates. For the Raman spectroscopy data, various preprocessing operations were systematically combined allowing the identification of a preprocessing pipeline, which proved to effectively eliminate initial lysate composition variations as well as most interferences attributed to precipitates and the precipitant present in solution. The calibrated partial least squares models seamlessly predicted the precipitant concentration with R 2 of 0.98 and 0.97 in batch and fed-batch experiments, respectively, and captured the observed precipitation trends with R 2 of 0.74 and 0.64. Although the resolution of fine differences between experiments was limited due to the observed non-linear relationship between spectral data and the VLP concentration, this study provides a foundation for employing Raman spectroscopy as a PAT sensor for monitoring VLP precipitation processes with the potential to extend its applicability to other phase-behavior dependent processes or molecules.

Automated classification pipeline for real-time in vivo examination of colorectal tissue using Raman spectroscopy

Colorectal cancer is the third most common malignancy worldwide and one of the leading causes of death in oncological patients with its diagnosis typically involving confirmation by tissue biopsy. In vivo Raman spectroscopy, an experimental diagnostic method less invasive than a biopsy, has shown great potential to discriminate between normal and cancerous tissue. However, the complex and often manual processing of Raman spectra along with the absence of a suitable instant classifier are the main obstacles to its adoption in clinical practice. This study aims to address these issues by developing a real-time automated classification pipeline coupled with a user-friendly application tailored for non-spectroscopists. First, in addition to routine colonoscopy, 377 subjects underwent in vivo acquisitions of Raman spectra of healthy tissue, adenomatous polyps, or cancerous tissue, which were conducted using a custom-made microprobe. The spectra were then loaded into the pipeline and pre-processed in several steps, including standard normal variate transformation and finite impulse response filtration. The quality of the pre-processed spectral data was checked based on their signal-to-noise ratio before the suitable spectra were decomposed and classified using a combination of principal component analysis and a support vector machine, respectively. After five-fold cross-validation, the developed classifier exhibited 100% sensitivity toward adenocarcinoma and adenomatous polyps. The overall accuracy was 96.9% and 79.2% for adenocarcinoma and adenomatous polyps respectively. In addition, an application with a graphical user interface was developed to facilitate the use of our data pipeline by medical professionals in a clinical environment. Overall, the combination of supervised and unsupervised machine learning with algorithmic pre-processing of in vivo Raman spectra appears to be a viable way of reducing the relatively large number of biopsies currently needed to definitively diagnose colorectal cancer.

Spectroscopic insights into multi-phase protein crystallization in complex lysate using Raman spectroscopy and a particle-free bypass

Protein crystallization as opposed to well-established chromatography processes has the benefits to reduce production costs while reaching a comparable high purity. However, monitoring crystallization processes remains a challenge as the produced crystals may interfere with analytical measurements. Especially for capturing proteins from complex feedstock containing various impurities, establishing reliable process analytical technology (PAT) to monitor protein crystallization processes can be complicated. In heterogeneous mixtures, important product characteristics can be found by multivariate analysis and chemometrics, thus contributing to the development of a thorough process understanding. In this project, an analytical set-up is established combining offline analytics, on-line ultraviolet visible light (UV/Vis) spectroscopy, and in-line Raman spectroscopy to monitor a stirred-batch crystallization process with multiple phases and species being present. As an example process, the enzyme Lactobacillus kefir alcohol dehydrogenase (LkADH) was crystallized from clarified Escherichia coli (E. coli) lysate on a 300 mL scale in five distinct experiments, with the experimental conditions changing in terms of the initial lysate solution preparation method and precipitant concentration. Since UV/Vis spectroscopy is sensitive to particles, a cross-flow filtration (cross-flow filtration)-based bypass enabled the on-line analysis of the liquid phase providing information on the lysate composition regarding the nucleic acid to protein ratio. A principal component analysis (PCA) of in situ Raman spectra supported the identification of spectra and wavenumber ranges associated with productspecific information and revealed that the experiments followed a comparable, spectral trend when crystals were present. Based on preprocessed Raman spectra, a partial least squares (PLS) regression model was optimized to monitor the target molecule concentration in real-time. The off-line sample analysis provided information on the crystal number and crystal geometry by automated image analysis as well as the concentration of LkADH and host cell proteins (HCPs) In spite of a complex lysate suspension containing scattering crystals and various impurities, it was possible to monitor the target molecule concentration in a heterogeneous, multi-phase process using spectroscopic methods. With the presented analytical set-up of off-line, particle-sensitive on-line, and in-line analyzers, a crystallization capture process can be characterized better in terms of the geometry, yield, and purity of the crystals.

Unveiling the efficacy of a bulk Raman spectra-based model in predicting single cell Raman spectra of microorganisms

In a previous publication, we trained predictive models based on Raman bulk spectra of microorganisms placed on a silicon dioxide protected silver mirror slide to make predictions for new Raman spectra, unknown to the models, of microorganisms placed on a different substrate, namely stainless steel. Now we have combined large sections of this data and trained a convolutional neural network (CNN) to make predictions for single cell Raman spectra. We show that a database based on microbial bulk material is conditionally suited to make predictions for the same species in terms of single cells. Data of 13 different microorganisms (bacteria and yeasts) were used. Two of the 13 species could be identified 90% correctly and five other species 71%-88%. The six remaining species were correctly predicted by only 0%-49%. Especially stronger fluorescence in bulk material compared to single cells but also photodegradation of carotenoids are some effects that can complicate predictions for single cells based on bulk data. The results could be helpful in assessing universal Raman tools or databases.

Siamese Networks for Clinically Relevant Bacteria Classification Based on Raman Spectroscopy

Identifying bacterial strains is essential in microbiology for various practical applications, such as disease diagnosis and quality monitoring of food and water. Classical machine learning algorithms have been utilized to identify bacteria based on their Raman spectra. However, convolutional neural networks (CNNs) offer higher classification accuracy, but they require extensive training sets and retraining of previous untrained class targets can be costly and time-consuming. Siamese networks have emerged as a promising solution. They are composed of two CNNs with the same structure and a final network that acts as a distance metric, converting the classification problem into a similarity problem. Classical machine learning approaches, shallow and deep CNNs, and two Siamese network variants were tailored and tested on Raman spectral datasets of bacteria. The methods were evaluated based on mean sensitivity, training time, prediction time, and the number of parameters. In this comparison, Siamese-model2 achieved the highest mean sensitivity of 83.61 ± 4.73 and demonstrated remarkable performance in handling unbalanced and limited data scenarios, achieving a prediction accuracy of 73%. Therefore, the choice of model depends on the specific trade-off between accuracy, (prediction/training) time, and resources for the particular application. Classical machine learning models and shallow CNN models may be more suitable if time and computational resources are a concern. Siamese networks are a good choice for small datasets and CNN for extensive data.

Label-free SERS detection of prostate cancer based on multi-layer perceptron surrogate model method

Prior surface-enhanced Raman spectroscopy (SERS) research has shown that pre-processing is necessary before analysis. Pre-processing also typically serves the dual purposes of removing the auto-fluorescence background and minimizing data volatility. This method allows for a more accurate comparison of spectral traits and relative SERS peak strength. However, because there are so many different kinds of samples, it can take a long time, and there is no assurance that the approach chosen will work well with a particular kind of sample. Therefore, this study employed a deep learning technique called multi-layer perceptron (MLP) to simplify the pre-processing of blood plasma SERS samples in patients with prostate cancer (PC), as well as to enhance the sensitivity and specificity of diagnosis using SERS technology. First of all, significant variations in peak intensity can be observed in the difference spectra, facilitating differentiation between PC and normal groups. Second, the data analysis was carried out in three different stages (raw data, defluorescenced data, and normalized data) using principal component analysis and linear discriminant analysis (PCA-LDA), as well as PCA-multi-layer perceptron (PCA-MLP). Finally, when SERS data was analyzed using PCA-LDA, there were significant differences in classification accuracy across each stage (The classification accuracy of three different stages were 76.90%, 85.60%, 95.20%, respectively). However, when PCA-MLP was utilized for SERS data analysis, the classification accuracy remained consistently high and stable (The classification accuracy of three different stages were 92.00%, 92.40%, 96.70%, respectively). The experimental results of PCA-MLP for classifying specific SERS data indicate that analyzing raw data directly can simplify the experimental process and enhance the efficacy of SERS analysis.

Multiplexing potential of NIR resonant and non-resonant Raman reporters for bio-imaging applications

Multiplexed imaging, which allows for the interrogation of multiple molecular features simultaneously, is vital for addressing numerous challenges across biomedicine. Optically unique surface-enhanced Raman scattering (SERS) nanoparticles (NPs) have the potential to serve as a vehicle to achieve highly multiplexed imaging in a single acquisition, which is non-destructive, quantitative, and simple to execute. When using laser excitation at 785 nm, which allows for a lower background from biological tissues, near infrared (NIR) dyes can be used as Raman reporters to provide high Raman signal intensity due to the resonance effect. This class of imaging agents are known as surface-enhanced resonance Raman scattering (SERRS) NPs. Investigators have predominantly utilized two classes of Raman reporters in their nanoparticle constructs for use in biomedical applications: NIR-resonant and non-resonant Raman reporters. Herein, we investigate the multiplexing potential of five non-resonant SERS: BPE, 44DP, PTT, PODT, and BMMBP, and five NIR resonant SERRS NP flavors with heptamethine cyanine dyes: DTTC, IR-770, IR-780, IR-792, and IR-797, which have been extensively used for biomedical imaging applications. Although SERRS NPs display high Raman intensities, due to their resonance properties, we observed that non-resonant SERS NP concentrations can be quantitated by the intensity of their unique emissions with higher accuracy. Spectral unmixing of five-plex mixtures revealed that the studied non-resonant SERS NPs maintain their detection limits more robustly as compared to the NIR resonant SERRS NP flavors when introducing more components into a mixture.

Assessment of Primary Human Liver Cancer Cells by Artificial Intelligence-Assisted Raman Spectroscopy

We investigated the possibility of using Raman spectroscopy assisted by artificial intelligence methods to identify liver cancer cells and distinguish them from their Non-Tumor counterpart. To this aim, primary liver cells (40 Tumor and 40 Non-Tumor cells) obtained from resected hepatocellular carcinoma (HCC) tumor tissue and the adjacent non-tumor area (negative control) were analyzed by Raman micro-spectroscopy. Preliminarily, the cells were analyzed morphologically and spectrally. Then, three machine learning approaches, including multivariate models and neural networks, were simultaneously investigated and successfully used to analyze the cells' Raman data. The results clearly demonstrate the effectiveness of artificial intelligence (AI)-assisted Raman spectroscopy for Tumor cell classification and prediction with an accuracy of nearly 90% of correct predictions on a single spectrum.

Exploring the Steps of Infrared (IR) Spectral Analysis: Pre-Processing, (Classical) Data Modelling, and Deep Learning

Infrared (IR) spectroscopy has greatly improved the ability to study biomedical samples because IR spectroscopy measures how molecules interact with infrared light, providing a measurement of the vibrational states of the molecules. Therefore, the resulting IR spectrum provides a unique vibrational fingerprint of the sample. This characteristic makes IR spectroscopy an invaluable and versatile technology for detecting a wide variety of chemicals and is widely used in biological, chemical, and medical scenarios. These include, but are not limited to, micro-organism identification, clinical diagnosis, and explosive detection. However, IR spectroscopy is susceptible to various interfering factors such as scattering, reflection, and interference, which manifest themselves as baseline, band distortion, and intensity changes in the measured IR spectra. Combined with the absorption information of the molecules of interest, these interferences prevent direct data interpretation based on the Beer-Lambert law. Instead, more advanced data analysis approaches, particularly artificial intelligence (AI)-based algorithms, are required to remove the interfering contributions and, more importantly, to translate the spectral signals into high-level biological/chemical information. This leads to the tasks of spectral pre-processing and data modeling, the main topics of this review. In particular, we will discuss recent developments in both tasks from the perspectives of classical machine learning and deep learning.

Halfway to Automated Feeding of Chinese Hamster Ovary Cells

This paper presents a comprehensive study on the development of models and soft sensors required for the implementation of the automated bioreactor feeding of Chinese hamster ovary (CHO) cells using Raman spectroscopy and chemometric methods. This study integrates various methods, such as partial least squares regression and variable importance in projection and competitive adaptive reweighted sampling, and highlights their effectiveness in overcoming challenges such as high dimensionality, multicollinearity and outlier detection in Raman spectra. This paper emphasizes the importance of data preprocessing and the relationship between independent and dependent variables in model construction. It also describes the development of a simulation environment whose core is a model of CHO cell kinetics. The latter allows the development of advanced control algorithms for nutrient dosing and the observation of the effects of different parameters on the growth and productivity of CHO cells. All developed models were validated and demonstrated to have a high robustness and predictive accuracy, which were reflected in a 40% reduction in the root mean square error compared to established methods. The results of this study provide valuable insights into the practical application of these methods in the field of monitoring and automated cell feeding and make an important contribution to the further development of process analytical technology in the bioprocess industry.

Factor analysis of the time series of SERS spectra reveals water arrangement and surface plasmon changes in Ag nanoparticle systems

The enhancement of Raman signals of molecules localized in the vicinity of plasmonic nanoparticles, known as surface-enhanced Raman scattering (SERS) effect, is strongly influenced by the selected excitation wavelength. The optimal excitation wavelength in SERS measurements is given by the position of the surface plasmon extinction (SPE) band of the studied system. Even a small change of the SPE band intensity, position and/or shape during the measurement may influence the SERS signal significantly. In this work, we prepared several systems of Ag nanoparticles, which were used for the demonstration how the information about SPE changes can be obtained by multivariate statistical analysis (factor analysis; FA) from SERS spectral sets, and employed in more precise and more comprehensive interpretation of the results. In non-aggregated Ag colloidal systems measured at the excitation wavelength of 445 nm, SPE band changes could be monitored by the analysis of water stretching vibration together with the vibrations in the fingerprint region. The FA of the water stretching band region was shown to provide unique information on both arrangement and disarrangement of water molecules in the vicinity of Ag NPs during the time evolution of these SERS active systems. In addition, the FA of the fingerprint region helped to monitor a rapid metalation of meso-tetrakis(N-methyl-4-pyridyl)porphine in etched SERS systems with Ag⁺ ions released from the NPs surface. In aggregated Ag colloidal systems measured at the excitation wavelength of 785 nm, the FA of SERS spectral sets enabled us to reveal the contribution of the 2nd electromagnetic enhancement to the overall SERS signal. The reliability of our conclusions was verified by comparing the results obtained from FA of SERS spectral sets with the data obtained from the parallel SPE measurements of the studied systems.

Raman spectroscopy in crop quality assessment: focusing on sensing secondary metabolites: a review

As a crop quality sensor, Raman spectroscopy has been consistently proposed as one of the most promising and non-destructive methods for qualitative and quantitative analysis of plant substances, because it can measure molecular structures in a short time without requiring pretreatment along with simple usage. The sensitivity of the Raman spectrum to target chemicals depends largely on the wavelength, intensity of the laser power, and exposure time. Especially for plant samples, it is very likely that the peak of the target material is covered by strong fluorescence effects. Therefore, methods using lasers with low energy causing less fluorescence, such as 785 nm or near-infrared, are vigorously discussed. Furthermore, advanced techniques for obtaining more sensitive and clear spectra, like surface-enhanced Raman spectroscopy, time-gated Raman spectroscopy or combination with thin-layer chromatography, are being investigated. Numerous interpretations of plant quality can be represented not only by the measurement conditions but also by the spectral analysis methods. Up to date, there have been attempted to optimize and generalize analysis methods. This review summarizes the state of the art of micro-Raman spectroscopy in crop quality assessment focusing on secondary metabolites, from in vitro to in vivo and even in situ, and suggests future research to achieve universal application.

An Optimization Approach Coupling Preprocessing with Model Regression for Enhanced Chemometrics

Chemometric methods are broadly used in the chemical and biochemical sectors. Typically, derivation of a regression model follows data preprocessing in a sequential manner. Yet, preprocessing can significantly influence the regression model and eventually its predictive ability. In this work, we investigate the coupling of preprocessing and model parameter estimation by incorporating them simultaneously in an optimization step. Common model selection techniques rely almost exclusively on the performance of some accuracy metric, yet having a quantitative metric for model robustness can prolong model up-time. Our approach is applied to optimize for model accuracy and robustness. This requires the introduction of a novel mathematical definition for robustness. We test our method in a simulated set up and with industrial case studies from multivariate calibration. The results highlight the importance of both accuracy and robustness properties and illustrate the potential of the proposed optimization approach toward automating the generation of efficient chemometric models.

Raman spectroscopy: Principles and recent applications in food safety

Food contaminant is a significant issue because of the adverse effects on human health and economy. Traditional detection methods such as liquid chromatography-mass spectroscopy for detecting food contaminants are expensive and time-consuming, and require highly-trained personnel and complicated sample pretreatment. Raman spectroscopy is an advanced analytical technique in a manner of non-destructive, rapid, cost-effective, and ultrasensitive sensing various hazards in agri-foods. In this chapter, we summarized the principle of Raman spectroscopy and surface enhanced Raman spectroscopy, the methods to process Raman spectra, the recent applications of Raman/SERS (surface-enhanced Raman spectroscopy) in detecting chemical contaminants (e.g., pesticides, antibiotics, mycotoxins, heavy metals, and food adulterants) and microbiological hazards (e.g., Salmonella, Campylobacter, Shiga toxigenic E. coli, Listeria, and Staphylococcus aureus) in foods.

Quantification of Drugs in Brain and Liver Mimetic Tissue Models Using Raman Spectroscopy

Quantitative analysis of drug delivery with in biological systems is an integral challenge in drug development. Analytical techniques are important for assessing both drug target delivery, target action, and drug toxicology. Using mimetic tissue models, we have investigated the efficacy of Raman spectroscopy in quantitative detection of alkyne group and deuterated drugs in rat brain and rat liver tissue models. Lasers with 671 nm and 785 nm wavelengths were assessed for their feasibility in this application due to opposing relative benefits and disadvantages. Thin tissue sections have been tested as a practical means of reducing autofluorescent background by minimizing out-of-focus tissue and therefore maximizing photobleaching rates. Alkyne-tagged drugs were quantitatively measured at 18 ± 5 μg/g drug/tissue mass ratio in rat brain and at 34 ± 6 μg/g in rat liver. Quantification calibration curves were generated for a range of concentrations from 0-500 μg/g. These results show the potential of Raman spectroscopy as a diffraction-limited spatially resolved imaging technique for assessing drug delivery in tissue applications.

Understanding Raman Spectral Based Classifications with Convolutional Neural Networks Using Practical Examples of Fungal Spores and Carotenoid-Pigmented Microorganisms

Numerous publications showing that robust prediction models for microorganisms based on Raman micro-spectroscopy in combination with chemometric methods are feasible, often with very precise predictions. Advances in machine learning and easier accessibility to software make it increasingly easy for users to generate predictive models from complex data. However, the question regarding why those predictions are so accurate receives much less attention. In our work, we use Raman spectroscopic data of fungal spores and carotenoid-containing microorganisms to show that it is often not the position of the peaks or the subtle differences in the band ratios of the spectra, due to small differences in the chemical composition of the organisms, that allow accurate classification. Rather, it can be characteristic effects on the baselines of Raman spectra in biochemically similar microorganisms that can be enhanced by certain data pretreatment methods or even neutral-looking spectral regions can be of great importance for a convolutional neural network. Using a method called Gradient-weighted Class Activation Mapping, we attempt to peer into the black box of convolutional neural networks in microbiological applications and show which Raman spectral regions are responsible for accurate classification.

Silver Nanostar-Based SERS for the Discrimination of Clinically Relevant Acinetobacter baumannii and Klebsiella pneumoniae Species and Clones

The development of rapid, reliable, and low-cost methods that enable discrimination among clinically relevant bacteria is crucial, with emphasis on those listed as WHO Global Priority 1 Critical Pathogens, such as carbapenem-resistant Acinetobacter baumannii and carbapenem-resistant or ESBL-producing Klebsiella pneumoniae. To address this problem, we developed and validated a protocol of surface-enhanced Raman spectroscopy (SERS) with silver nanostars for the discrimination of A. baumannii and K. pneumoniae species, and their globally disseminated and clinically relevant antibiotic resistant clones. Isolates were characterized by mixing bacterial colonies with silver nanostars, followed by deposition on filter paper for SERS spectrum acquisition. Spectral data were processed with unsupervised and supervised multivariate data analysis methods, including principal component analysis (PCA) and partial least-squares discriminant analysis (PLSDA), respectively. Our proposed SERS procedure using silver nanostars adsorbed to the bacteria, followed by multivariate data analysis, enabled differentiation between and within species. This pilot study demonstrates the potential of SERS for the rapid discrimination of clinically relevant A. baumannii and K. pneumoniae species and clones, displaying several advantages such as the ease of silver nanostars synthesis and the possible use of a handheld spectrometer, which makes this approach ideal for point-of-care applications.

Building an ensemble learning model for gastric cancer cell line classification via rapid raman spectroscopy

Cell misuse and cross-contamination can affect the accuracy of cell research results and result in wasted time, manpower and material resources. Thus, cell line identification is important and necessary. At present, the commonly used cell line identification methods need cell staining and culturing. There is therefore a need to develop a new method for the rapid and automated identification of cell lines. Raman spectroscopy has become one of the emerging techniques in the field of microbial identification, with the advantages of being rapid and noninvasive and providing molecular information for biological samples, which is beneficial in the identification of cell lines. In this study, we built a library of Raman spectra for gastric mucosal epithelial cell lines GES-1 and gastric cancer cell lines, such as AGS, BGC-823, HGC-27, MKN-45, MKN-74 and SNU-16. Five spectral datasets were constructed using spectral data and included the full spectrum, fingerprint region, high-wavelength number region and Raman background of Raman spectra. A stacking ensemble learning model, SL-Raman, was built for different datasets, and gastric cancer cell identification was achieved. For the gastric cancer cells we studied, the differentiation accuracy of SL-Raman was 100% for one of the gastric cancer cells and 100% for six of the gastric cancer cells. Additionally, the separation accuracy for two gastric cancer cells with different degrees of differentiation was 100%. These results demonstrate that Raman spectroscopy combined with SL-Raman may be a new method for the rapid and accurate identification of gastric cancer. In addition, the accuracy of 94.38% for classifying Raman spectral background data using machine learning demonstrates that the Raman spectral background contains some useful spectral features. These data have been overlooked in previous studies.

Identification of Bacterial Pathogens at Genus and Species Levels through Combination of Raman Spectrometry and Deep-Learning Algorithms

The rapid and accurate identification of the causing agents during bacterial infections would greatly improve pathogen transmission, prevention, patient care, and medical treatments in clinical settings. Although many conventional and molecular methods have been proven to be efficient and reliable, some of them suffer technical biases and limitations that require the development and application of novel and advanced techniques. Recently, due to its cost affordability, noninvasiveness, and label-free feature, Raman spectroscopy (RS) is emerging as a potential technique for fast bacterial detection. However, the method is still hampered by many technical issues, such as low signal intensity, poor reproducibility, and standard data set insufficiency, among others. Thus, it should be cautiously claimed that Raman spectroscopy could provide practical applications in real-world settings. In order to evaluate the implementation potentials of Raman spectroscopy in the identification of bacterial pathogens, we investigated 30 bacterial species belonging to 9 different bacterial genera that were isolated from clinical samples via surfaced enhanced Raman spectroscopy (SERS). A total of 17,149 SERS spectra were harvested from a Raman spectrometer and were further analyzed via machine learning approaches, which showed that a convolutional neural network (CNN) deep learning algorithm achieved the highest prediction accuracy for recognizing pathogenic bacteria at both the genus and species levels. In summary, the SERS technique holds a promising potential for fast bacterial pathogen identification in clinical laboratories with the integration of machine learning algorithms, which might be further developed and sharpened for the direct identification and prediction of bacterial pathogens from clinical samples. IMPORTANCE In this study, we investigated 30 bacterial species belonging to 9 different bacterial genera that were isolated from clinical samples via surfaced enhanced Raman spectroscopy (SERS). A total of 17,149 SERS spectra were harvested from a Raman spectrometer and were further analyzed via machine learning approaches, the results of which showed that the convolutional neural network (CNN) deep learning algorithm could achieve the highest prediction accuracy for recognizing pathogenic bacteria at both the genus and species levels. Taken together, we concluded that the SERS technique held a promising potential for fast bacterial pathogen diagnosis in clinical laboratories with the integration of deep learning algorithms, which might be further developed and sharpened for the direct identification and prediction of bacterial pathogens from clinical samples.

Deep-learning framework for fully-automated recognition of TiO2 polymorphs based on Raman spectroscopy

Emerging machine learning techniques can be applied to Raman spectroscopy measurements for the identification of minerals. In this project, we describe a deep learning-based solution for automatic identification of complex polymorph structures from their Raman signatures. We propose a new framework using Convolutional Neural Networks and Long Short-Term Memory networks for compound identification. We train and evaluate our model using the publicly-available RRUFF spectral database. For model validation purposes, we synthesized and identified different TiO₂ polymorphs to evaluate the performance and accuracy of the proposed framework. TiO₂ is a ubiquitous material playing a crucial role in many industrial applications. Its unique properties are currently used advantageously in several research and industrial fields including energy storage, surface modifications, optical elements, electrical insulation to microelectronic devices such as logic gates and memristors. The results show that our model correctly identifies pure Anatase and Rutile with a high degree of confidence. Moreover, it can also identify defect-rich Anatase and modified Rutile based on their modified Raman Spectra. The model can also correctly identify the key component, Anatase, from the P25 Degussa TiO₂. Based on the initial results, we firmly believe that implementing this model for automatically detecting complex polymorph structures will significantly increase the throughput, while dramatically reducing costs.

Recent advances in muscle food safety evaluation: Hyperspectral imaging analyses and applications

As there is growing interest in process control for quality and safety in the meat industry, by integrating spectroscopy and imaging technologies into one system, hyperspectral imaging, or chemical or spectroscopic imaging has become an alternative analytical technique that can provide the spatial distribution of spectrum for fast and nondestructive detection of meat safety. This review addresses the configuration of the hyperspectral imaging system and safety indicators of muscle foods involving biological, chemical, and physical attributes and other associated hazards or poisons, which could cause safety problems. The emphasis focuses on applications of hyperspectral imaging techniques in the safety evaluation of muscle foods, including pork, beef, lamb, chicken, fish and other meat products. Although HSI can provide the spatial distribution of spectrum, characterized by overtones and combinations of the C-H, N-H, and O-H groups using different combinations of a light source, imaging spectrograph and camera, there still needs improvement to overcome the disadvantages of HSI technology for further applications at the industrial level.

Chemometrics for Raman Spectroscopy Harmonization

Raman spectroscopy is used in a wide variety of fields, and in a plethora of different configurations. Raman spectra of simple analytes can often be analyzed using univariate approaches and interpreted in a straightforward manner. For more complex spetral data such as time series or line profiles (1D), Raman maps (2D), or even volumes (3D), multivariate data analysis (MVDA) becomes a requirement. Even though there are some existing standards for creation, implementation, and validation of methods and models employed in industry and academics, further research and development in the field must contribute to their improvement. This review will cover, in broad terms, existing techniques as well as new developments for MVDA for Raman spectroscopic data, and in particular the use associated with instrumentation and data calibration. Chemometric models are often generated via fusion of analytical data from different sources, which enhances model discrimination and prediction abilities as compared to models derived from a single data source. For Raman spectroscopy, raw or unprocessed data is rarely ever used. Instead, spectra are usually corrected and manipulated,¹ often by case-specific rather than universal methods. Calibration models can be used to characterize qualitatively and/or quantitatively samples measured with the same instrumentation that was used to create the model. However, regular validation is required to ensure that aging or incorrect maintenance of the instrument does not alter the model's predictions, particularly when applied in regulated fields such as pharmaceuticals. Furthermore, a model transfer may be required for different reasons, such as replacement or significant repair of the instrumentation. Modeling can also be used to consistently harmonize Raman spectroscopic data across several instrumental designs, accounting for variations in the resulting spectrum induced by different components. Data for Raman harmonization models should be processed in a protocolled manner, and the original data accessible to allow for model reconstruction or transfer when new data is added. Important processing steps will be the calibration of the spectral axes and instrument dependent effects, such as spectral resolution. In addition, data fusion and model transfer are essential for allowing new instrumentation to build on existing models to harmonize their own data. Ideally, an open access database would be created and maintained, for the purpose of allowing for continued harmonization of new Raman instruments using an outlined and accepted protocol.

Raman spectroscopy and machine learning for the classification of esophageal squamous carcinoma

Early diagnosis of esophageal squamous cell carcinoma (ESCC), a common malignant tumor with a low overall survival rate due to metastasis and recurrence, is critical for effective treatment and improved prognosis. Raman spectroscopy, an advanced detection technology for esophageal cancer, was developed to improve diagnosis sensitivity, specificity, and accuracy. This study proposed a novel, effective, and noninvasive Raman spectroscopy technique to differentiate and classify ESCC cell lines. Seven ESCC cell lines and tissues of an ESCC patient with staging of T3N1M0 and T3N2M0 at low and high differentiation levels were investigated through Raman spectroscopy. Raman spectral data analysis was performed with four machine learning algorithms, namely principal components analysis (PCA)- linear discriminant analysis (LDA), PCA-eXtreme gradient boosting (XGB), PCA- support vector machine (SVM), and PCA- (LDA, XGB, SVM)-stacked Gradient Boosting Machine (GBM). Four machine learning algorithms were able to classifiy ESCC cell subtypes from normal esophageal cells. The PCA-XGB model achieved an overall predictive accuracy of 85% for classifying ESCC and adjacent tissues. Moreover, an overall predictive accuracy of 90.3% was achieved in distinguishing low differentiation and high differentiation ESCC tissues with the same stage when PCA-LDA, XGM, and SVM models were combined. This study illustrated the Raman spectral traits of ESCC cell lines and esophageal tissues related to clinical pathological diagnosis. Future studies should investigate the role of Raman spectral features in ESCC pathogenesis.

Machine Learning of Raman Spectroscopy Data for Classifying Cancers: A Review of the Recent Literature

Raman Spectroscopy has long been anticipated to augment clinical decision making, such as classifying oncological samples. Unfortunately, the complexity of Raman data has thus far inhibited their routine use in clinical settings. Traditional machine learning models have been used to help exploit this information, but recent advances in deep learning have the potential to improve the field. However, there are a number of potential pitfalls with both traditional and deep learning models. We conduct a literature review to ascertain the recent machine learning methods used to classify cancers using Raman spectral data. We find that while deep learning models are popular, and ostensibly outperform traditional learning models, there are many methodological considerations which may be leading to an over-estimation of performance; primarily, small sample sizes which compound sub-optimal choices regarding sampling and validation strategies. Amongst several recommendations is a call to collate large benchmark Raman datasets, similar to those that have helped transform digital pathology, which researchers can use to develop and refine deep learning models.

Marker-Free, Molecule Sensitive Mapping of Disturbed Falling Fluid Films Using Raman Imaging

Technical liquid flow films are the basic arrangement for gas fluid transitions of all kinds and are the basis of many chemical processes, such as columns, evaporators, dryers, and different other kinds of fluid/fluid separation units. This publication presents a new method for molecule sensitive, non-contact, and marker-free localized concentration mapping in vertical falling films. Using Raman spectroscopy, no label or marker is needed for the detection of the local composition in liquid mixtures. In the presented cases, the film mapping of sodium sulfate in water on a plain surface as well as an added artificial streaming disruptor with the shape of a small pyramid is scanned in three dimensions. The results show, as a prove of concept, a clear detectable spectroscopic difference between air, back plate, and sodium sulfate for every local point in all three dimensions. In conclusion, contactless Raman scanning on falling films for liquid mapping is realizable without any mechanical film interaction caused by the measuring probe. Surface gloss or optical reflections from a metallic back plate are suppressed by using only inelastic light scattering and the mathematical removal of background noise.

Featuring few essential Raman spectroscopic signatures between heterogeneous cells

Here we demonstrate it is instructive to quantify cell Raman spectroscopy by sparse regularization. To be able to extract the specific spectral differences in a heterogeneous cell system with great spectroscopic similarities derived from many common biomolecular components, the maximum information entropy probability was proposed and exemplified by identifying normal lymphocytes from leukemia cells. The essential spectroscopic features were observed to locate at three Raman peaks whose spectral signatures were commensurate. The applicability of the extracted features was acknowledged by that the predicted identification accuracy of up to 93% was still achieved when only two peaks were loaded into decision tree model, which may provide the possibility of a clinically rapid hematological malignancy detection.

An antigen-targeting assay for Lyme disease: Combining aptamers and SERS to detect the OspA protein

Lyme disease is the fastest growing vector-borne disease in the United States. However, current testing modalities are ill suited to detection of Lyme disease, leading to the diagnosis of many cases after treatment is effective. We present an improved, direct method Lyme disease diagnosis, where the Lyme specific biomarker Outer Surface Protein A (OspA) in clinical serum samples is identified using a diagnostic platform combining surface enhanced Raman scattering (SERS) and aptamers. Employing orthogonal projections to latent structures discriminant analysis, the system accurately identified 91% of serum samples from Lyme patients, and 96% of serum samples from symptomatic controls. In addition, the OspA limit-of-detection, determined to be 1 × 10^-4 ng/mL, is greater than four orders of magnitude lower than that found in serum samples from early Lyme disease patients. The application of this platform to detect this difficult-to-diagnose disease suggests its potential for detecting other diseases that present similar difficulties.

Optimised Pre-Processing of Raman Spectra for Colorectal Cancer Detection Using High-Performance Computing

Spectral pre-processing is an essential step in data analysis for biomedical diagnostic applications of Raman spectroscopy, allowing the removal of undesirable spectral contributions that could mask biological information used for diagnosis. However, due to the specificity of pre-processing for a given sample type and the vast number of potential pre-processing combinations, optimisation of pre-processing via a manual "trial and error" format is often time intensive with no guarantee that the chosen method is optimal for the sample type. Here we present the use of high-performance computing (HPC) to trial over 2.4 million pre-processing permutations to demonstrate the optimisation on the pre-processing of human serum Raman spectra for colorectal cancer detection. The effect of varying pre-processing order, using extended multiplicative scatter correction, spectral smoothing, baseline correction, binning and normalization was considered. Permutations were assessed on their ability to detect patients with disease using a random forest (RF) algorithm trained with 102 patients (510 spectra) and independently tested with a set of 439 patients (1317 spectra) in a primary care patient cohort. Optimising via HPC enables improved performance in diagnostic abilities, with sensitivity increasing by 14.6%, specificity increasing by 6.9%, positive predictive value increasing by 3.4%, and negative predictive value increasing by 2.4% when compared to a standard pre-processing optimisation. Ultimate values of these metrics are very important for diagnostic adoption, and once diagnostics demonstrate good accuracy these types of optimisations can make a significant difference to roll-out of a test and demonstrating advantages over existing tests. We also provide tips/recommendations for pre-processing optimisation without the use of HPC. From the HPC permutations, recommendations for appropriate parameter constraints for conducting a more basic pre-processing optimisation are also detailed, thus helping model development for researchers not having access to HPC.

Development of distribution maps of spectrally similar degradation products by Raman chemical imaging microscope coupled with a new variable selection technique and SIMCA classifier

The ability to detect degradation products of active pharmaceutical ingredients (API) is an essential performance not only for conducting proper stability studies and subsequently gain regulatory approvals; but as well for detecting degradation products during the manufacturing process (In Process Control). Thus, this study aims to present the ability of using Raman Chemical Imaging (Raman-CI) microscope, with its optimum precision, in combination with appropriate chemometrics algorithms, to detect the spectrally similar Salicylic Acid (SA) in Acetylsalicylic Acid (ASA) powder mixture, and then create a chemical distribution map that reflects the distribution of ASA's main degradation product. The generated Hyperspectral images were processed where, a supervised chemometrics soft classifier, Soft Independent Modeling of Class Analogy (SIMCA), is applied to classify pixels and construct the subsequent distribution maps. In addition, due to the challenge of the high structural and spectral similarity between both substances, this study presents a new variable selection and dimensionality reduction technique, called Variable Strength Coefficient (VSC) to maximize the spectral differences and enhance the model precision and selectivity. A High-performance liquid chromatographic (HPLC) method was applied as a reference separation method to assess the results obtained by the proposed technique. The proposed technique was validated, where the obtained results confirmed that Raman Chemical Imaging Microscope, when coupled with SIMCA and VSC, is a powerful tool with outstanding accuracy. In addition, this approach could be suitable in applications where constructing accurate distribution maps of spectrally similar API's is required.

Nondestructive 3D imaging and quantification of hydrated biofilm matrix by confocal Raman microscopy coupled with non-negative matrix factorization

Biofilms are ubiquitous in natural and engineered environments and of great importance in drinking water distribution and biological wastewater treatment systems. Simultaneously acquiring the chemical and structural information of the hydrated biofilm matrix is essential for the cognition and regulation of biofilms in the environmental field. However, the complexity of samples and the limited approaches prevent a holistic understanding of the biofilm matrix. In this work, an approach based on the confocal Raman mapping technique integrated with non-negative matrix factorization (NMF) analysis was developed to probe the hydrated biofilm matrix in situ. The flexibility of the NMF analysis was utilized to subtract the undesired water background signal and resolve the meaningful biological components from Raman spectra of the hydrated biofilms. Diverse chemical components such as proteins, bacterial cells, glycolipids and polyhydroxyalkanoates (PHA) were unraveled within the distinct Pseudomonas spp. biofilm matrices, and the corresponding 3-dimensional spatial organization was visualized and quantified. Of these components, glycolipids and PHA were unique to the P. aeruginosa and P. putida biofilm matrix, respectively. Furthermore, their high abundances in the lower region of the biofilm matrix were found to be related to the specific physiological functions and surrounding microenvironments. Overall, the results demonstrate that our NMF Raman mapping method could serve as a powerful tool complementary to the conventional approaches for identifying and visualizing the chemical components in the biofilm matrix. This work may facilitate the online characterization of the biofilm matrix widely present in the environment and advance the fundamental understanding of biofilm.

Distinguishing cotton seed genotypes by means of vibrational spectroscopic methods (NIR and Raman) and chemometrics

The use of vibrational spectroscopy, such as near infrared (NIR) and Raman, combined with multivariate analysis methods to analyze agricultural products are promising for investigating genetically modified organisms (GMO). In Brazil, cotton is grown under humid tropical conditions and is highly affected by pests and diseases, requiring the use of large amounts of phytosanitary chemicals. To avoid the use of those pesticides, genetic improvement can be carried out to produce species tolerant to herbicides, resistant to fungi and insects, or even to provide greater productivity and better quality. Even with these advantages, it is necessary to manage and limit the contact of transgenic species with native ones, avoiding possible contamination or even extinction of conventional species. The identification of the presence of GMOs is based on complex DNA-based analysis, which is usually laborious, expensive, time-consuming, destructive, and generally unavailable. In the present study, a new methodology to identify GMOs using partial least squares discriminant analysis (PLS-DA) on NIR and Raman data is proposed to distinguish conventional and transgenic cotton seed genotypes, providing classification errors for prediction set of 2.23% for NIR and 0.0% for Raman.

Compositional assessment of bone by Raman spectroscopy

Raman spectroscopy (RS) is used to analyze the physiochemical properties of bone because it is non-destructive and requires minimal sample preparation. With over two decades of research involving measurements of mineral-to-matrix ratio, type-B carbonate substitution, crystallinity, and other compositional characteristics of the bone matrix by RS, there are multiple methods to acquire Raman signals from bone, to process those signals, and to determine peak ratios including sub-peak ratios as well as the full-width at half maximum of the most prominent Raman peak, which is nu1 phosphate (ν1PO4). Selecting which methods to use is not always clear. Herein, we describe the components of RS instruments and how they influence the quality of Raman spectra acquired from bone because signal-to-noise of the acquisition and the accompanying background fluorescence dictate the pre-processing of the Raman spectra. We also describe common methods and challenges in preparing acquired spectra for the determination of matrix properties of bone. This article also serves to provide guidance for the analysis of bone by RS with examples of how methods for pre-processing the Raman signals and for determining properties of bone composition affect RS sensitivity to potential differences between experimental groups. Attention is also given to deconvolution methods that are used to ascertain sub-peak ratios of the amide I band as a way to assess characteristics of collagen type I. We provide suggestions and recommendations on the application of RS to bone with the goal of improving reproducibility across studies and solidify RS as a valuable technique in the field of bone research.

Chemometric analysis in Raman spectroscopy from experimental design to machine learning-based modeling

Raman spectroscopy is increasingly being used in biology, forensics, diagnostics, pharmaceutics and food science applications. This growth is triggered not only by improvements in the computational and experimental setups but also by the development of chemometric techniques. Chemometric techniques are the analytical processes used to detect and extract information from subtle differences in Raman spectra obtained from related samples. This information could be used to find out, for example, whether a mixture of bacterial cells contains different species, or whether a mammalian cell is healthy or not. Chemometric techniques include spectral processing (ensuring that the spectra used for the subsequent computational processes are as clean as possible) as well as the statistical analysis of the data required for finding the spectral differences that are most useful for differentiation between, for example, different cell types. For Raman spectra, this analysis process is not yet standardized, and there are many confounding pitfalls. This protocol provides guidance on how to perform a Raman spectral analysis: how to avoid these pitfalls, and strategies to circumvent problematic issues. The protocol is divided into four parts: experimental design, data preprocessing, data learning and model transfer. We exemplify our workflow using three example datasets where the spectra from individual cells were collected in single-cell mode, and one dataset where the data were collected from a raster scanning-based Raman spectral imaging experiment of mice tissue. Our aim is to help move Raman-based technologies from proof-of-concept studies toward real-world applications.

First preliminary study on identification of bacterial fish pathogens with Raman spectroscopy

Accurate and rapid determination of bacterial disease agents of fish is an important step for sustainable and efficient aquaculture production. In general, biochemical and molecular methods are used for pathogen detection but they are usually time-consuming and required qualified personnel. Recently spectroscopic methods are preferred in clinical and food microbiology and declared as a promising alternative method for pathogens diagnosis with many advantages. In this study, the significant spectra of three important bacterial fish pathogens (Lactococcus garvieae, Vibrio anguillarum and Yersinia ruckeri) were determined by Raman spectroscopy. The first data of the pathogens were obtained and the distinctive differences in polysaccharides, nucleic acids, fatty acids and amino acids were identified. This preliminary study aimed to be pioneer for further studies in aquaculture and veterinary microbiology toward developing an alternative method for routine identification.

A polyyne toxin produced by an antagonistic bacterium blinds and lyses a Chlamydomonad alga

Algae live in association with microbes that interact by a variety of chemical mediators, resulting in mutualistic or antagonistic relationships. Although algae are key contributors to carbon fixation and are fundamental for food webs, we still know little about the underlying molecular mechanisms affecting their fitness. This study investigates the interaction between an antagonistic bacterium and a unicellular alga. It demonstrates multiple roles of a polyyne, protegencin, that is used by the bacteria to attack green algal cells. It is a highly effective toxin that alters a subcellular algal compartment used for vision, bleaches, and lyses the algal cells. These results expand our knowledge of the arsenal of chemical mediators in bacteria and their modes of action in algal communities.

Algae are key contributors to global carbon fixation and form the basis of many food webs. In nature, their growth is often supported or suppressed by microorganisms. The bacterium Pseudomonas protegens Pf-5 arrests the growth of the green unicellular alga Chlamydomonas reinhardtii, deflagellates the alga by the cyclic lipopeptide orfamide A, and alters its morphology [P. Aiyar et al., Nat. Commun. 8, 1756 (2017)]. Using a combination of Raman microspectroscopy, genome mining, and mutational analysis, we discovered a polyyne toxin, protegencin, which is secreted by P. protegens, penetrates the algal cells, and causes destruction of the carotenoids of their primitive visual system, the eyespot. Together with secreted orfamide A, protegencin thus prevents the phototactic behavior of C. reinhardtii. A mutant of P. protegens deficient in protegencin production does not affect growth or eyespot carotenoids of C. reinhardtii. Protegencin acts in a direct and destructive way by lysing and killing the algal cells. The toxic effect of protegencin is also observed in an eyeless mutant and with the colony-forming Chlorophyte alga Gonium pectorale. These data reveal a two-pronged molecular strategy involving a cyclic lipopeptide and a conjugated tetrayne used by bacteria to attack select Chlamydomonad algae. In conjunction with the bloom-forming activity of several chlorophytes and the presence of the protegencin gene cluster in over 50 different Pseudomonas genomes [A. J. Mullins et al., bioRxiv [Preprint] (2021). https://www.biorxiv.org/content/10.1101/2021.03.05.433886v1 (Accessed 17 April 2021)], these data are highly relevant to ecological interactions between Chlorophyte algae and Pseudomonadales bacteria.

Investigating microcrystalline cellulose crystallinity using Raman spectroscopy

Microcrystalline cellulose (MCC) is a semi-crystalline material with inherent variable crystallinity due to raw material source and variable manufacturing conditions. MCC crystallinity variability can result in downstream process variability. The aim of this study was to develop models to determine MCC crystallinity index (%CI) from Raman spectra of 30 commercial batches using Raman probes with spot sizes of 100 µm (MR probe) and 6 mm (PhAT probe). A principal component analysis model separated Raman spectra of the same samples captured using the different probes. The %CI was determined using a previously reported univariate model based on the ratio of the peaks at 380 and 1096 cm^-1. The univariate model was adjusted for each probe. The %CI was also predicted from spectral data from each probe using partial least squares regression models (where Raman spectra and univariate %CI were the dependent and independent variables, respectively). Both models showed adequate predictive power. For these models a general reference amorphous spectrum was proposed for each instrument. The development of the PLS model substantially reduced the analysis time as it eliminates the need for spectral deconvolution. A web application containing all the models was developed.

SUPPLEMENTARY INFORMATION

The online version contains supplementary material available at 10.1007/s10570-021-04093-1.

Morpho-Molecular Metabolic Analysis and Classification of Human Pituitary Gland and Adenoma Biopsies Based on Multimodal Optical Imaging

Pituitary adenomas count among the most common intracranial tumors. During pituitary oncogenesis structural, textural, metabolic and molecular changes occur which can be revealed with our integrated ultrahigh-resolution multimodal imaging approach including optical coherence tomography (OCT), multiphoton microscopy (MPM) and line scan Raman microspectroscopy (LSRM) on an unprecedented cellular level in a label-free manner. We investigated 5 pituitary gland and 25 adenoma biopsies, including lactotroph, null cell, gonadotroph, somatotroph and mammosomatotroph as well as corticotroph. First-level binary classification for discrimination of pituitary gland and adenomas was performed by feature extraction via radiomic analysis on OCT and MPM images and achieved an accuracy of 88%. Second-level multi-class classification was performed based on molecular analysis of the specimen via LSRM to discriminate pituitary adenomas subtypes with accuracies of up to 99%. Chemical compounds such as lipids, proteins, collagen, DNA and carotenoids and their relation could be identified as relevant biomarkers, and their spatial distribution visualized to provide deeper insight into the chemical properties of pituitary adenomas. Thereby, the aim of the current work was to assess a unique label-free and non-invasive multimodal optical imaging platform for pituitary tissue imaging and to perform a multiparametric morpho-molecular metabolic analysis and classification.

Design of a Multimodal Imaging System and Its First Application to Distinguish Grey and White Matter of Brain Tissue. A Proof-of-Concept-Study

Multimodal imaging gains increasing popularity for biomedical applications. This article presents the design of a novel multimodal imaging system. The centerpiece is a light microscope operating in the incident and transmitted light mode. Additionally, Raman spectroscopy and VIS/NIR reflectance spectroscopy are adapted. The proof-of-concept is realized to distinguish between grey matter (GM) and white matter (WM) of normal mouse brain tissue. Besides Raman and VIS/NIR spectroscopy, the following optical microscopy techniques are applied in the incident light mode: brightfield, darkfield, and polarization microscopy. To complement the study, brightfield images of a hematoxylin and eosin (H&E) stained cryosection in the transmitted light mode are recorded using the same imaging system. Data acquisition based on polarization microscopy and Raman spectroscopy gives the best results regarding the tissue differentiation of the unstained section. In addition to the discrimination of GM and WM, both modalities are suited to highlight differences in the density of myelinated axons. For Raman spectroscopy, this is achieved by calculating the sum of two intensity peak ratios (I2857 + I2888)/I2930 in the high-wavenumber region. For an optimum combination of the modalities, it is recommended to apply the molecule-specific but time-consuming Raman spectroscopy to smaller regions of interest, which have previously been identified by the microscopic modes.

Rapid detection and discrimination of food-related bacteria using IR-microspectroscopy in combination with multivariate statistical analysis

Spoilage microorganisms are of great concern for the food industry. While traditional culturing methods for spoilage microorganism detection are laborious and time-consuming, the development of early detection methods has gained a lot of interest in the last decades. In this work a rapid and non-destructive detection and discrimination method of eight important food-related microorganisms (Bacillus subtilis DSM 10, Bacillus coagulans DSM 1, Escherichia coli K12 DSM 498, Escherichia coli TOP10, Micrococcus luteus DSM 20030, Pseudomonas fluorescens DSM 4358, Pseudomonas fluorescens DSM 50090 and Bacillus thuringiensis israelensis DSM 5724) based on IR-microspectroscopy and chemometric evaluation was developed. Sampling was carried out directly from the surface to be tested, without the need for sample preparation such as purification, singulation, centrifugation and washing steps, as an efficient and inexpensive blotting technique using the sample carrier. IR spectra were recorded directly after the blotting from the surface of the sample carrier without any further pretreatments. A combination of data preprocessing, principal component analysis and canonical discriminant analysis was found to be suitable. The spectral range from 400 to 1750 cm^-1 of the IR-microspectrosopic data was determined to be highly sensitive to the time after incubation and sample thickness, resulting in a high standard deviation. Therefore, this area was excluded from the evaluation in favor of the meaningfulness of the chemometric model and, thus, only the spectral range of specific -CH/-NH/-OH excitations (2815-3680 cm^-1) was used for model development. This study showed that the differentiation of food-related microorganisms on genera, species and strain level is feasible. A leave-one-out cross-validation of the training data set showed 100% accuracy. The classification of the ungrouped test data showed with an accuracy of 94.5% that, despite the large biological variance of the analytes such as different times after incubation and the presented sampling (including its variance), a robust and meaningful model for the differentiation of food-related bacteria could be developed by data preprocessing and subsequent chemometric evaluation.

Vibrational Spectroscopic Investigation of Blood Plasma and Serum by Drop Coating Deposition for Clinical Application

In recent decades, vibrational spectroscopic methods such as Raman and FT-IR spectroscopy are widely applied to investigate plasma and serum samples. These methods are combined with drop coating deposition techniques to pre-concentrate the biomolecules in the dried droplet to improve the detected vibrational signal. However, most often encountered challenge is the inhomogeneous redistribution of biomolecules due to the coffee-ring effect. In this study, the variation in biomolecule distribution within the dried-sample droplet has been investigated using Raman and FT-IR spectroscopy and fluorescence lifetime imaging method. The plasma-sample from healthy donors were investigated to show the spectral differences between the inner and outer-ring region of the dried-sample droplet. Further, the preferred location of deposition of the most abundant protein albumin in the blood during the drying process of the plasma has been illustrated by using deuterated albumin. Subsequently, two patients with different cardiac-related diseases were investigated exemplarily to illustrate the variation in the pattern of plasma and serum biomolecule distribution during the drying process and its impact on patient-stratification. The study shows that a uniform sampling position of the droplet, both at the inner and the outer ring, is necessary for thorough clinical characterization of the patient's plasma and serum sample using vibrational spectroscopy.