Analysis of Raman Spectra by Using Deep Learning Methods in the Identification of Marine Pathogens

RamanCluster: A deep clustering-based framework for unsupervised Raman spectral identification of pathogenic bacteria

Raman spectroscopy serves as a powerful and reliable tool for the characterization of pathogenic bacteria. The integration of Raman spectroscopy with artificial intelligence techniques to rapidly identify pathogenic bacteria has become paramount for expediting disease diagnosis. However, the development of prevailing supervised artificial intelligence algorithms is still constrained by costly and limited well-annotated Raman spectroscopy datasets. Furthermore, tackling various high-dimensional and intricate Raman spectra of pathogenic bacteria in the absence of annotations remains a formidable challenge. In this paper, we propose a concise and efficient deep clustering-based framework (RamanCluster) to achieve accurate and robust unsupervised Raman spectral identification of pathogenic bacteria without the need for any annotated data. RamanCluster is composed of a novel representation learning module and a machine learning-based clustering module, systematically enabling the extraction of robust discriminative representations and unsupervised Raman spectral identification of pathogenic bacteria. The extensive experimental results show that RamanCluster has achieved high accuracy on both Bacteria-4 and Bacteria-6, with ACC values of 77 % and 74.1 %, NMI values of 75 % and 73 %, as well as AMI values of 74.6 % and 72.6 %, respectively. Furthermore, compared with other state-of-the-art methods, RamanCluster exhibits the superior accuracy on handling various complicated pathogenic bacterial Raman spectroscopy datasets, including situations with strong noise and a wide variety of pathogenic bacterial species. Additionally, RamanCluster also demonstrates commendable robustness in these challenging scenarios. In short, RamanCluster has a promising prospect in accelerating the development of low-cost and widely applicable disease diagnosis in clinical medicine.

Microstrip isoelectric focusing with deep learning for simultaneous screening of diabetes, anemia, and thalassemia

BACKGROUND

Hemoglobin (Hb) is an important protein in red blood cells and a crucial diagnostic indicator of diseases, e.g., diabetes, thalassemia, and anemia. However, there is a rare report on methods for the simultaneous screening of diabetes, anemia, and thalassemia. Isoelectric focusing (IEF) is a common separative tool for the separation and analysis of Hb. However, the current analysis of IEF images is time-consuming and cannot be used for simultaneous screening. Therefore, an artificial intelligence (AI) of IEF image recognition is desirable for accurate, sensitive, and low-cost screening.

RESULTS

Herein, we proposed a novel comprehensive method based on microstrip isoelectric focusing (mIEF) for detecting the relative content of Hb species. There was a good coincidence between the quantitation of Hb via a conventional automated hematology analyzer and the one via mIEF with R2 = 0.9898. Nevertheless, our results showed that the accuracy of disease diagnosis based on the quantification of Hb species alone is as low as 69.33 %, especially for the simultaneous screening of multiple diseases of diabetes, anemia, alpha-thalassemia, and beta-thalassemia. Therefore, we introduced a ResNet1D-based diagnosis model for the improvement of screening accuracy of multiple diseases. The results showed that the proposed model could achieve a high accuracy of more than 90 % and a good sensitivity of more than 96 % for each disease, indicating the overwhelming advantage of the mIEF method combined with deep learning in contrast to the pure mIEF method.

SIGNIFICANCE

Overall, the presented method of mIEF with deep learning enabled, for the first time, the absolute quantitative detection of Hb, relative quantitation of Hb species, and simultaneous screening of diabetes, anemia, alpha-thalassemia, and beta-thalassemia. The AI-based diagnosis assistant system combined with mIEF, we believe, will help doctors and specialists perform fast and precise disease screening in the future.

Applications of Data Characteristic AI-Assisted Raman Spectroscopy in Pathological Classification

Raman spectroscopy has been widely used for label-free biomolecular analysis of cells and tissues for pathological diagnosis in vitro and in vivo. AI technology facilitates disease diagnosis based on Raman spectroscopy, including machine learning (PCA and SVM), manifold learning (UMAP), and deep learning (ResNet and AlexNet). However, it is not clear how to optimize the appropriate AI classification model for different types of Raman spectral data. Here, we selected five representative Raman spectral data sets, including endometrial carcinoma, hepatoma extracellular vesicles, bacteria, melanoma cell, diabetic skin, with different characteristics regarding sample size, spectral data size, Raman shift range, tissue sites, Kullback-Leibler (KL) divergence, and significant Raman shifts (i.e., wavenumbers with significant differences between groups), to explore the performance of different AI models (e.g., PCA-SVM, SVM, UMAP-SVM, ResNet or AlexNet). For data set of large spectral data size, Resnet performed better than PCA-SVM and UMAP. By building data characteristic-assisted AI classification model, we optimized the network parameters (e.g., principal components, activation function, and loss function) of AI model based on data size and KL divergence etc. The accuracy improved from 85.1 to 94.6% for endometrial carcinoma grading, from 77.1 to 90.7% for hepatoma extracellular vesicles detection, from 89.3 to 99.7% for melanoma cell detection, from 88.1 to 97.9% for bacterial identification, from 53.7 to 85.5% for diabetic skin screening, and mean time expense of 5 s.

Combined Mutual Learning Net for Raman Spectral Microbial Strain Identification

Infectious diseases pose a significant threat to global health, yet traditional microbiological identification methods suffer from drawbacks, such as high costs and long processing times. Raman spectroscopy, a label-free and noninvasive technique, provides rich chemical information and has tremendous potential in fast microbial diagnoses. Here, we propose a novel Combined Mutual Learning Net that precisely identifies microbial subspecies. It demonstrated an average identification accuracy of 87.96% in an open-access data set with thirty microbial strains, representing a 5.76% improvement. 50% of the microbial subspecies accuracies were elevated by 1% to 46%, especially for E. coli 2 improved from 31% to 77%. Furthermore, it achieved a remarkable subspecies accuracy of 92.4% in the custom-built fiber-optical tweezers Raman spectroscopy system, which collects Raman spectra at a single-cell level. This advancement demonstrates the effectiveness of this method in microbial subspecies identification, offering a promising solution for microbiology diagnosis.

A Comparative Analysis of Data Synthesis Techniques to Improve Classification Accuracy of Raman Spectroscopy Data

Raman spectra are examples of high dimensional data that can often be limited in the number of samples. This is a primary concern when Deep Learning frameworks are developed for tasks such as chemical species identification, quantification, and diagnostics. Open-source data are difficult to obtain and often sparse; furthermore, the collecting and curating of new spectra require expertise and resources. Deep generative modeling utilizes Deep Learning architectures to approximate high dimensional distributions and aims to generate realistic synthetic data. The evaluation of the data and the performance of the deep models is usually conducted on a per-task basis and provides no indication of an increase to robustness, or generalization, on a wider scale. In this study, we compare the benefits and limitations of a standard statistical approach to data synthesis (weighted blending) with a popular deep generative model, the Variational Autoencoder. Two binary data sets are divided into 3-fold to simulate small, limited samples. Synthetic data distributions are created per fold using the two methods and then augmented into the training of two Deep Learning algorithms, a Convolutional Neural Network and a Fully-Connected Neural Network. The goal of this study is to observe the trends in learning as synthetic data are continually augmented to the training data in increasing batches. To determine the impact of each synthetic method, Principal Component Analysis and the discrete Fréchet distance are implemented to visualize and measure the distance between the source and synthetic distributions along with the Machine Learning metric balanced accuracy for evaluating performance on imbalanced data.

Rapid identification of lactic acid bacteria at species/subspecies level via ensemble learning of Ramanomes

Rapid and accurate identification of lactic acid bacteria (LAB) species would greatly improve the screening rate for functional LAB. Although many conventional and molecular methods have proven efficient and reliable, LAB identification using these methods has generally been slow and tedious. Single-cell Raman spectroscopy (SCRS) provides the phenotypic profile of a single cell and can be performed by Raman spectroscopy (which directly detects vibrations of chemical bonds through inelastic scattering by a laser light) using an individual live cell. Recently, owing to its affordability, non-invasiveness, and label-free features, the Ramanome has emerged as a potential technique for fast bacterial detection. Here, we established a reference Ramanome database consisting of SCRS data from 1,650 cells from nine LAB species/subspecies and conducted further analysis using machine learning approaches, which have high efficiency and accuracy. We chose the ensemble meta-classifier (EMC), which is suitable for solving multi-classification problems, to perform in-depth mining and analysis of the Ramanome data. To optimize the accuracy and efficiency of the machine learning algorithm, we compared nine classifiers: LDA, SVM, RF, XGBoost, KNN, PLS-DA, CNN, LSTM, and EMC. EMC achieved the highest average prediction accuracy of 97.3% for recognizing LAB at the species/subspecies level. In summary, Ramanomes, with the integration of EMC, have promising potential for fast LAB species/subspecies identification in laboratories and may thus be further developed and sharpened for the direct identification and prediction of LAB species from fermented food.

Illuminating the Tiny World: A Navigation Guide for Proper Raman Studies on Microorganisms

Raman spectroscopy is an emerging method for the identification of bacteria. Nevertheless, a lot of different parameters need to be considered to establish a reliable database capable of identifying real-world samples such as medical or environmental probes. In this review, the establishment of such reliable databases with the proper design in microbiological Raman studies is demonstrated, shining a light into all the parts that require attention. Aspects such as the strain selection, sample preparation and isolation requirements, the phenotypic influence, measurement strategies, as well as the statistical approaches for discrimination of bacteria, are presented. Furthermore, the influence of these aspects on spectra quality, result accuracy, and read-out are discussed. The aim of this review is to serve as a guide for the design of microbiological Raman studies that can support the establishment of this method in different fields.

Patch-Based Convolutional Encoder: A Deep Learning Algorithm for Spectral Classification Balancing the Local and Global Information

Molecular vibrational spectroscopies, including infrared absorption and Raman scattering, provide molecular fingerprint information and are powerful tools for qualitative and quantitative analysis. They benefit from the recent development of deep-learning-based algorithms to improve the spectral, spatial, and temporal resolutions. Although a variety of deep-learning-based algorithms, including those to simultaneously extract the global and local spectral features, have been developed for spectral classification, the classification accuracy is still far from satisfactory when the difference becomes very subtle. Here, we developed a lightweight algorithm named patch-based convolutional encoder (PACE), which effectively improved the accuracy of spectral classification by extracting spectral features while balancing local and global information. The local information was captured well by segmenting the spectrum into patches with an appropriate patch size. The global information was extracted by constructing the correlation between different patches with depthwise separable convolutions. In the five open-source spectral data sets, PACE achieved a state-of-the-art performance. The more difficult the classification, the better the performance of PACE, compared with that of residual neural network (ResNet), vision transformer (ViT), and other commonly used deep learning algorithms. PACE helped improve the accuracy to 92.1% in Raman identification of pathogen-derived extracellular vesicles at different physiological states, which is much better than those of ResNet (85.1%) and ViT (86.0%). In general, the precise recognition and extraction of subtle differences offered by PACE are expected to facilitate vibrational spectroscopy to be a powerful tool toward revealing the relevant chemical reaction mechanisms in surface science or realizing the early diagnosis in life science.

Rapid, label-free classification of glioblastoma differentiation status combining confocal Raman spectroscopy and machine learning

Label-free identification of tumor cells using spectroscopic assays has emerged as a technological innovation with a proven ability for rapid implementation in clinical care. Machine learning facilitates the optimization of processing and interpretation of extensive data, such as various spectroscopy data obtained from surgical samples. The here-described preclinical work investigates the potential of machine learning algorithms combining confocal Raman spectroscopy to distinguish non-differentiated glioblastoma cells and their respective isogenic differentiated phenotype by means of confocal ultra-rapid measurements. For this purpose, we measured and correlated modalities of 1146 intracellular single-point measurements and sustainingly clustered cell components to predict tumor stem cell existence. By further narrowing a few selected peaks, we found indicative evidence that using our computational imaging technology is a powerful approach to detect tumor stem cells in vitro with an accuracy of 91.7% in distinct cell compartments, mainly because of greater lipid content and putative different protein structures. We also demonstrate that the presented technology can overcome intra- and intertumoral cellular heterogeneity of our disease models, verifying the elevated physiological relevance of our applied disease modeling technology despite intracellular noise limitations for future translational evaluation.

Semi-Supervised Deep Learning-Based Multi-component Spectral Calibration Modeling for UV-vis and Near-Infrared Spectroscopy without Information Loss

Spectral analysis is an important method for characterizing and identifying chemical species. However, quantitative spectral analysis of multiple chemical properties in the real world has always been a challenging problem due to the strong correlation, massive noise, and serious information overlapping of the spectral features. Here, we present a new semi-supervised spectral calibration method based on information lossless decoupling of spectral features named NICEM. To realize the separation and extraction of key latent features, the method uses the flow-based model non-linear independent component estimation (NICE) to learn the sample distribution. The spectral data information is transformed into independent latent variables obeying Gaussian distribution by the reversible structure of deep network without information loss, so as to find the essential properties and realize the feature nonlinear decomposition. Moreover, the association between the input latent feature variables and attributes is evaluated by the maximum mutual information coefficient to eliminate the adverse effects of irrelevant information in the latent variable space and mine key information. Since the latent variables are independent in each dimension, the NICEM method is easier to establish an accurate semi-supervised multi-component calibration model even for high overlapping and complex spectral data. The applicability of the proposed spectral modeling method is demonstrated by using three ultraviolet-visible and near-infrared spectral data sets with 15 physical and chemical properties including diesel fuels, corn, and multi-metal ions solution. Results show that the proposed NICEM method has the highest determination coefficient (R2) and significantly improves extrapolation compared with the seven state-of-the-art methods. The proposed method is intuitive because it obviates complex feature engineering and prior knowledge and is a promising spectral calibration tool for quantitative analysis in other spectroscopy applications.

Multi-scale sequential feature selection for disease classification using Raman spectroscopy data

Raman spectroscopy (RS) optical technology promises non-destructive and fast application in medical disease diagnosis in a single step. However, achieving clinically relevant performance levels remains challenging due to the inability to search for significant Raman signals at different scales. Here we propose a multi-scale sequential feature selection method that can capture global sequential features and local peak features for disease classification using RS data. Specifically, we utilize the Long short-term memory network (LSTM) module to extract global sequential features in the Raman spectra, as it can capture long-term dependencies present in the Raman spectral sequences. Meanwhile, the attention mechanism is employed to select local peak features that were ignored before and are the key to distinguishing different diseases. Experimental results on three public and in-house datasets demonstrate the superiority of our model compared with state-of-the-art methods for RS classification. In particular, our model achieves an accuracy of 97.9 ± 0.2% on the COVID-19 dataset, 76.3 ± 0.4% on the H-IV dataset, and 96.8 ± 1.9% on the H-V dataset.

A microfluidic approach for label-free identification of small-sized microplastics in seawater

Marine microplastics are emerging as a growing environmental concern due to their potential harm to marine biota. The substantial variations in their physical and chemical properties pose a significant challenge when it comes to sampling and characterizing small-sized microplastics. In this study, we introduce a novel microfluidic approach that simplifies the trapping and identification process of microplastics in surface seawater, eliminating the need for labeling. We examine various models, including support vector machine, random forest, convolutional neural network (CNN), and residual neural network (ResNet34), to assess their performance in identifying 11 common plastics. Our findings reveal that the CNN method outperforms the other models, achieving an impressive accuracy of 93% and a mean area under the curve of 98 ± 0.02%. Furthermore, we demonstrate that miniaturized devices can effectively trap and identify microplastics smaller than 50 µm. Overall, this proposed approach facilitates efficient sampling and identification of small-sized microplastics, potentially contributing to crucial long-term monitoring and treatment efforts.

A Universal and Accurate Method for Easily Identifying Components in Raman Spectroscopy Based on Deep Learning

Raman spectroscopy has been widely used to provide the structural fingerprint for molecular identification. Due to interference from coexisting components, noise, baseline, and systematic differences between spectrometers, component identification with Raman spectra is challenging, especially for mixtures. In this study, a method entitled DeepRaman has been proposed to solve those problems by combining the comparison ability of a pseudo-Siamese neural network (pSNN) and the input-shape flexibility of spatial pyramid pooling (SPP). DeepRaman was trained, validated, and tested with 41,564 augmented Raman spectra from two databases (pharmaceutical material and S.T. Japan). It can achieve 96.29% accuracy, 98.40% true positive rate (TPR), and 94.36% true negative rate (TNR) on the test set. Another six data sets measured on different instruments were used to evaluate the performance of the proposed method from different aspects. DeepRaman can provide accurate identification results and significantly outperform the hit quality index (HQI) method and other deep learning models. In addition, it performs well in cases of different spectral complexity and low-content components. Once the model is established, it can be used directly on different data sets without retraining or transfer learning. Furthermore, it also obtains promising results for the analysis of surface-enhanced Raman spectroscopy (SERS) data sets and Raman imaging data sets. In summary, it is an accurate, universal, and ready-to-use method for component identification in various application scenarios.

Accurate identification of living Bacillus spores using laser tweezers Raman spectroscopy and deep learning

Accurately, rapidly, and noninvasively identifying Bacillus spores can greatly contribute to controlling a plenty of infectious diseases. Laser tweezers Raman spectroscopy (LTRS) has confirmed to be a powerful tool for studying Bacillus spores at a single cell level. In this study, we constructed a single-cell Raman spectra dataset of living Bacillus spores and utilized deep learning approach to accurately, nondestructively identify Bacillus spores. The trained convolutional neural network (CNN) could efficiently extract tiny Raman spectra features of five spore species, and provide a prediction accuracy of specie identification as high as 100 %. Moreover, the spectral feature differences in three Raman bands at 660, 826, and 1017 cm^-1 were confirmed to mostly contribute to producing such high prediction accuracy. In addition, optimal CNN model was employed to monitor and identify sporulation process at different metabolic phases in one growth cycle. The obtained average prediction accuracy of metabolic phase identification was approximately 88 %. It can be foreseen that, LTRS combined with CNN approach have great potential for accurately identifying spore species and metabolic phases at a single cell level, and can be gradually extended to perform identification for many unculturable bacteria growing in soil, water, and food.

A one-dimensional convolutional neural network based deep learning for high accuracy classification of transformation stages in esophageal squamous cell carcinoma tissue using micro-FTIR

Among the most frequently diagnosed cancers in developing countries, esophageal squamous cell carcinoma (ESCC) ranks among the top six causes of death. It would be beneficial if a rapid, accurate, and automatic ESCC diagnostic method could be developed to reduce the workload of pathologists and improve the effectiveness of cancer treatments. Using micro-FTIR spectroscopy, this study classified the transformation stages of ESCC tissues. Based on 6,352 raw micro-FTIR spectra, a one-dimensional convolutional neural network (1D-CNN) model was constructed to classify-five stages. Based on the established model, more than 93% accuracy was achieved at each stage, and the accuracy of identifying proliferation, low grade neoplasia, and ESCC cancer groups was achieved 99% for the test dataset. In this proof-of-concept study, the developed method can be applied to other diseases in order to promote the use of FTIR spectroscopy in cancer pathology.

Profiling of Tumor Cell-Delivered Exosome by Surface Enhanced Raman Spectroscopy-Based Biosensor for Evaluation of Nasopharyngeal Cancer Radioresistance

Although the advancement of radiotherapy significantly improves the survival of nasopharyngeal cancer (NPC), radioresistance associated with recurrence and poor outcomes still remains a daunting challenge in the clinical scenario. Currently, effective biomarkers and convenient detection methods for predicting radioresistance have not been well established. Here, the surface-enhanced Raman spectroscopy combined with proteomics is used to firstly profile the characteristic spectral patterns of exosomes secreted from self-established NPC radioresistance cells, and reveals specific variations of proteins expression during radioresistance formation, including collagen alpha-2 (I) chain (COL1A2) that is associated with a favorable prognosis in NPC and is negatively associated with DNA repair scores and DNA repair-related genes via bioinformatic analysis. Furthermore, deep learning model-based diagnostic model is generated to accurately identify the exosomes from radioresistance group. This work demonstrates the promising potential of exosomes as a novel biomarker for predicting the radioresistance and develops a rapid and sensitive liquid biopsy method that will provide a personalized and precise strategy for clinical NPC treatment.

Nontargeted detection and recognition of adulterants in milk powder using Raman imaging and neural networks

Raman imaging technology combined with targeted chemometrics can play a vital role in the rapid detection of milk powder adulteration, which threatens the lives of infants and other people. However, these methods always suffer from a narrow detection range. Nontargeted methods show a broader detection range but cannot recognize adulterants. Here, a novel nontargeted chemometric method, named as the adversarial discrimination neural network (ADNN), is proposed to detect and recognize adulterants simultaneously. The method comprises building a tight boundary in the feature space of Raman images to discriminate milk powder samples from the majority of adulterated cases. Then a first-order partial derivative of the ADNN is calculated to recognize different adulterants through a local approximation strategy. A validation set containing samples adulterated with various adulterants at concentrations ranging from 0.3% to 1.5% w/w was provided to challenge the proposed method. The validated detection accuracy of the proposed method for authentic and adulterated samples was 99.9% and 99.7% and the adulterants were recognized correctly. The ADNN-Raman represents a novel nontargeted and end-to-end tool for detecting and recognizing adulterants in milk powder simultaneously, providing new insights into nontargeted chemometric analysis.

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.

Could artificial intelligence revolutionize the development of nanovectors for gene therapy and mRNA vaccines?

Gene therapy enables the introduction of nucleic acids like DNA and RNA into host cells, and is expected to revolutionize the treatment of a wide range of diseases. This growth has been further accelerated by the discovery of CRISPR/Cas technology, which allows accurate genomic editing in a broad range of cells and organisms in vitro and in vivo. Despite many advances in gene delivery and the development of various viral and non-viral gene delivery vectors, the lack of highly efficient non-viral systems with low cellular toxicity remains a challenge. The application of cutting-edge technologies such as artificial intelligence (AI) has great potential to find new paradigms to solve this issue. Herein, we review AI and its major subfields including machine learning (ML), neural networks (NNs), expert systems, deep learning (DL), computer vision and robotics. We discuss the potential of AI-based models and algorithms in the design of targeted gene delivery vehicles capable of crossing extracellular and intracellular barriers by viral mimicry strategies. We finally discuss the role of AI in improving the function of CRISPR/Cas systems, developing novel nanobots, and mRNA vaccine carriers.

Bisphenol A and microplastics weaken the antimicrobial ability of blood clams by disrupting humoral immune responses and suppressing hemocyte chemotactic activity

Robust antimicrobial capability is crucial for marine organisms survival in complex ocean environments. Although the detrimental impacts of emergent pollutants on cellular immune response of marine bivalve mollusks were increasingly documented, the effects of bisphenol A (BPA) and microplastics (MPs) on humoral immune response and hemocyte chemotactic activity remain unclear. Therefore, in this study, the toxicities of BPA and MPs, alone or in combination, to the antimicrobial ability, humoral immune response, and hemocyte chemotactic activity were investigated in the blood clam Tegillarca granosa. Our data demonstrated that exposure of blood clams to BPA, MPs, and BPA-MPs for 2 weeks lead to significant reductions in their survival rates upon pathogenic bacterial challenge, indicating evident impairment of antimicrobial ability. Compared to control, the plasma of pollutant-incubated blood clams exhibited significantly less antimicrobial activity against the growth of V. harveyi, suggesting significant reduction in humoral immune effectors including defensin, lysozyme (LZM), and lectin. Moreover, hemocytes migration across the polycarbonate membrane to the serum containing chamber was markedly arrested by 2-week exposure to BPA, MPs, and BPA-MPs, suggesting a hampered chemotactic activity. In addition, the intracellular contents of ROS and protein carbonyl in hemocytes were markedly induced whereas the expression levels of key genes from the MAPK and actin cytoskeleton regulation pathways were significantly suppressed upon exposure. In this study, it was also found that BPA-MP coexposure was significantly more toxic than single exposures. In summary, our findings revealed that exposure to the pollutants tested possibly impair the antimicrobial ability of blood clam through (1) reducing the inhibitory effect of plasma on bacterial growth, the contents of humoral immune effectors, and the chemotactic activity of hemocytes, (2) interrupting IL-17 activation of MAPK signal pathway, (3) inducing intracellular ROS, elevating protein carbonylation levels, and disrupting actin cytoskeleton regulation in hemocytes.

Visualization of a Machine Learning Framework toward Highly Sensitive Qualitative Analysis by SERS

Surface-enhanced Raman spectroscopy (SERS), providing near-single-molecule-level fingerprint information, is a powerful tool for the trace analysis of a target in a complicated matrix and is especially facilitated by the development of modern machine learning algorithms. However, both the high demand of mass data and the low interpretability of the mysterious black-box operation significantly limit the well-trained model to real systems in practical applications. Aiming at these two issues, we constructed a novel machine learning algorithm-based framework (Vis-CAD), integrating visual random forest, characteristic amplifier, and data augmentation. The introduction of data augmentation significantly reduced the requirement of mass data, and the visualization of the random forest clearly presented the captured features, by which one was able to determine the reliability of the algorithm. Taking the trace analysis of individual polycyclic aromatic hydrocarbons in a mixture as an example, a trustworthy accuracy no less than 99% was realized under the optimized condition. The visualization of the algorithm framework distinctly demonstrated that the captured feature was well correlated to the characteristic Raman peaks of each individual. Furthermore, the sensitivity toward the trace individual could be improved by least 1 order of magnitude as compared to that with the naked eye. The proposed algorithm distinguished by the lesser demand of mass data and the visualization of the operation process offers a new way for the indestructible application of machine learning algorithms, which would bring push-to-the-limit sensitivity toward the qualitative and quantitative analysis of trace targets, not only in the field of SERS, but also in the much wider spectroscopy world. It is implemented in the Python programming language and is open-source at https://github.com/3331822w/Vis-CAD.

In situ identification of environmental microorganisms with Raman spectroscopy

Microorganisms in natural environments are crucial in maintaining the material and energy cycle and the ecological balance of the environment. However, it is challenging to delineate environmental microbes' actual metabolic pathways and intraspecific heterogeneity because most microorganisms cannot be cultivated. Raman spectroscopy is a culture-independent technique that can collect molecular vibration profiles from cells. It can reveal the physiological and biochemical information at the single-cell level rapidly and non-destructively in situ. The first part of this review introduces the principles, advantages, progress, and analytical methods of Raman spectroscopy applied in environmental microbiology. The second part summarizes the applications of Raman spectroscopy combined with stable isotope probing (SIP), fluorescence in situ hybridization (FISH), Raman-activated cell sorting and genomic sequencing, and machine learning in microbiological studies. Finally, this review discusses expectations of Raman spectroscopy and future advances to be made in identifying microorganisms, especially for uncultured microorganisms.

Classification of skin cancer using convolutional neural networks analysis of Raman spectra

BACKGROUND AND OBJECTIVE

Skin cancer is the most common malignancy in whites accounting for about one third of all cancers diagnosed per year. Portable Raman spectroscopy setups for skin cancer "optical biopsy" are utilized to detect tumors based on their spectral features caused by the comparative presence of different chemical components. However, low signal-to-noise ratio in such systems may prevent accurate tumors classification. Thus, there is a challenge to develop methods for efficient skin tumors classification.

METHODS

We compare the performance of convolutional neural networks and the projection on latent structures with discriminant analysis for discriminating skin cancer using the analysis of Raman spectra with a high autofluorescence background stimulated by a 785 nm laser. We have registered the spectra of 617 cases of skin neoplasms (615 patients, 70 melanomas, 122 basal cell carcinomas, 12 squamous cell carcinomas and 413 benign tumors) in vivo with a portable Raman setup and created classification models both for convolutional neural networks and projection on latent structures approaches. To check the classification models stability, a 10-fold cross-validation was performed for all created models. To avoid models overfitting, the data was divided into a training set (80% of spectral dataset) and a test set (20% of spectral dataset).

RESULTS

The results for different classification tasks demonstrate that the convolutional neural networks significantly (p<0.01) outperforms the projection on latent structures. For the convolutional neural networks implementation we obtained ROC AUCs of 0.96 (0.94 - 0.97; 95% CI), 0.90 (0.85-0.94; 95% CI), and 0.92 (0.87 - 0.97; 95% CI) for classifying a) malignant vs benign tumors, b) melanomas vs pigmented tumors and c) melanomas vs seborrheic keratosis respectively.

CONCLUSIONS

The performance of the convolutional neural networks classification of skin tumors based on Raman spectra analysis is higher or comparable to the accuracy provided by trained dermatologists. The increased accuracy with the convolutional neural networks implementation is due to a more precise accounting of low intensity Raman bands in the intense autofluorescence background. The achieved high performance of skin tumors classifications with convolutional neural networks analysis opens a possibility for wide implementation of Raman setups in clinical setting.

Research on the Comprehensive Evaluation Method for the Automatic Recognition of Raman Spectrum under Multidimensional Constraint

Raman spectrum contains abundant substance information with fingerprint characteristics. However, due to the huge variety of substances and their complex characteristic information, it is difficult to recognize the Raman spectrum accurately. Starting from dimensions like the Raman shift, the relative peak intensity, and the overall hit ratio of characteristic peaks, we extracted and recognized the characteristics in the Raman spectrum and analyzed these characteristics from local and global perspectives and then proposed a comprehensive evaluation method for the recognition of Raman spectrum on the basis of the data fusion of the recognition results under multidimensional constraint. Based on the common spectrum database of the normal Raman and surface-enhanced Raman of thousands of substances, we analyzed the performance of the evaluation method. It shows that even for the identification of spectra from instruments of low technical specifications, the automatic recognition rate of the sample can reach 98% and above, a great improvement compared with that of the common identification algorithms, which proves the effectiveness of the comprehensive evaluation method under multidimensional constraint.

Extreme Point Sort Transformation Combined With a Long Short-Term Memory Network Algorithm for the Raman-Based Identification of Therapeutic Monoclonal Antibodies

Therapeutic monoclonal antibodies (mAbs) are a new generation of protein-based medicines that are usually expensive and thus represent a target for counterfeiters. In the present study, a method based on Raman spectroscopy that combined extreme point sort transformation with a long short-term memory (LSTM) network algorithm was presented for the identification of therapeutic mAbs. A total of 15 therapeutic mAbs were used in this study. An in-house Raman spectrum dataset for model training was created with 1,350 spectra. The characteristic region of the Raman spectrum was reduced in dimension and then transformed through an extreme point sort transformation into a sequence array, which was fitted for the LSTM network. The characteristic array was extracted from the sequence array using a well-trained LSTM network and then compared with standard spectra for identification. To demonstrate whether the present algorithm was better, ThermoFisher OMNIC 8.3 software (Thermo Fisher Scientific Inc., U.S.) with two matching modes was selected for comparison. Finally, the present method was successfully applied to identify 30 samples, including 15 therapeutic mAbs and 15 other injections. The characteristic region was selected from 100 to 1800 cm⁻¹ of the full spectrum. The optimized dimensional values were set from 35 to 53, and the threshold value range was from 0.97 to 0.99 for 15 therapeutic mAbs. The results of the robustness test indicated that the present method had good robustness against spectral peak drift, random noise and fluorescence interference from the measurement. The areas under the curve (AUC) values of the present method that were analysed on the full spectrum and analysed on the characteristic region by the OMNIC 8.3 software’s built-in method were 1.000, 0.678, and 0.613, respectively. The similarity scores for 15 therapeutic mAbs using OMNIC 8.3 software in all groups compared with that of the relative present algorithm group had extremely remarkable differences (p < 0.001). The results suggested that the extreme point sort transformation combined with the LSTM network algorithm enabled the characteristic extraction of the therapeutic mAb Raman spectrum. The present method is a proposed solution to rapidly identify therapeutic mAbs.

Development of deep learning algorithms to discriminate giant cell tumors of bone from adjacent normal tissues by confocal Raman spectroscopy

Raman spectroscopy is a non-destructive analysis technique that provides detailed information about the chemical structure of tumors. Raman spectra of 52 giant cell tumors of bone (GCTB) and 21 adjacent normal tissues of formalin-fixed paraffin embedded (FFPE) and frozen specimens were obtained using a confocal Raman spectrometer and analyzed with machine learning and deep learning algorithms. We discovered characteristic Raman shifts in the GCTB specimens. They were assigned to phenylalanine and tyrosine. Based on the spectroscopic data, classification algorithms including support vector machine, k-nearest neighbors and long short-term memory (LSTM) were successfully applied to discriminate GCTB from adjacent normal tissues of both the FFPE and frozen specimens, with the accuracy ranging from 82.8% to 94.5%. Importantly, our LSTM algorithm showed the best performance in the discrimination of the frozen specimens, with a sensitivity and specificity of 93.9% and 95.1% respectively, and the AUC was 0.97. The results of our study suggest that confocal Raman spectroscopy accomplished by the LSTM network could non-destructively evaluate a tumor margin by its inherent biochemical specificity which may allow intraoperative assessment of the adequacy of tumor clearance.

Raman Microscopic Identification of Microorganisms on Metal Surfaces via Support Vector Machines

An easy, inexpensive, and rapid method to identify microorganisms is in great demand in various areas such as medical diagnostics or in the food industry. In our study, we show the development of several predictive models based on Raman spectroscopy combined with support vector machines (SVM) for 21 species of microorganisms. The microorganisms, grown under standardized conditions, were placed on a silver mirror slide to record the data for model development. Additional data was obtained from microorganisms on a polished stainless-steel slide in order to validate the models in general and to assess possible negative influences of the material change on the predictions. The theoretical prediction accuracies for the most accurate models, based on a five-fold cross-validation, are 98.4%. For practical validation, new spectra (from stainless-steel surfaces) have been used, which were not included in the calibration data set. The overall prediction accuracy in practice was about 80% and the inaccurate predictions were only due to a few species. The development of a database provides the basis for further investigations such as the application and extension to single-cell analytics and for the characterization of biofilms.

A Review of Raman-Based Technologies for Bacterial Identification and Antimicrobial Susceptibility Testing

Antimicrobial resistance (AMR) is a global medical threat that seriously endangers human health. Rapid bacterial identification and antimicrobial susceptibility testing (AST) are key interventions to combat the spread and emergence of AMR. Although current clinical bacterial identification and AST provide comprehensive information, they are labor-intensive, complex, inaccurate, and slow (requiring several days, depending on the growth of pathogenic bacteria). Recently, Raman-based identification and AST technologies have played an increasingly important role in fighting AMR. This review summarizes major Raman-based techniques for bacterial identification and AST, including spontaneous Raman scattering, surface-enhanced Raman scattering (SERS), and coherent Raman scattering (CRS) imaging. Then, we discuss recent developments in rapid identification and AST methods based on Raman technology. Finally, we highlight the major challenges and potential future efforts to improve clinical outcomes through rapid bacterial identification and AST.

Raman spectrum matching with contrastive representation learning

An effective contrastive representation learning method for spectra identification with a frequentist guarantee of including the correct class prediction on two Raman datasets (Mineral and Organic) and one SERS dataset (Bacteria).