Search

This study presents a novel method for the diagnosis of hepatitis B virus infection using human blood serum Raman spectroscopy combined with a deep learning model. The sera of 499 people infected with the hepatitis B virus and 435 healthy controls were measured in this experiment. The data were subjected to a dimensionality reduction by principal component analysis. Then, the features of multiple scales were preserved and fused by a multiscale fusion convolution operation. The gated recurrent unit network was added to extract time series features and finally output the result of the classification through a softmax layer. A diagnostic model based on a gated recurrent unit and multiscale fusion convolutional neural network was constructed and evaluated by a 10-fold cross-validation method. Compared to existing analysis methods for serum Raman spectroscopy, the proposed model achieved the best performance. The combination of Raman spectroscopy and deep learning models is expected to be applied well in the early screening of hepatitis B and is a promising screening method. [ABSTRACT FROM AUTHOR]

Celiac Disease (CD) is a primary malabsorption syndrome resulting from the interplay of genetic, immune, and dietary factors. CD negatively impacts daily activities and may lead to conditions such as osteoporosis, malignancies in the small intestine, ulcerative jejunitis, and enteritis, ultimately causing severe malnutrition. Therefore, an effective and rapid differentiation between healthy individuals and those with celiac disease is crucial for early diagnosis and treatment. This study utilizes Raman spectroscopy combined with deep learning models to achieve a non-invasive, rapid, and accurate diagnostic method for celiac disease and healthy controls. A total of 59 plasma samples, comprising 29 celiac disease cases and 30 healthy controls, were collected for experimental purposes. Convolutional Neural Network (CNN), Multi-Scale Convolutional Neural Network (MCNN), Residual Network (ResNet), and Deep Residual Shrinkage Network (DRSN) classification models were employed. The accuracy rates for these models were found to be 86.67%, 90.76%, 86.67% and 95.00%, respectively. Comparative validation results revealed that the DRSN model exhibited the best performance, with an AUC value and accuracy of 97.60% and 95%, respectively. This confirms the superiority of Raman spectroscopy combined with deep learning in the diagnosis of celiac disease. [ABSTRACT FROM AUTHOR]

As a highly prevalent and recurrent cancer, detecting the degree of differentiation in oral cancer is crucial. Current methods rely on biopsies in the presence of significant lesions, which are time-consuming. This study introduces an efficient and rapid approach for mass determination of oral cancer differentiation levels. By leveraging serum Raman spectroscopy combined with deep neural networks and extreme gradient boosting, we performed feature selection and interaction on oral cancer samples of varying differentiation levels, achieving the most accurate and reliable classification of oral cancer differentiation stages. [ABSTRACT FROM AUTHOR]

The aim of this study was to explore the feasibility of Raman spectroscopy combined with computer algorithms in the diagnosis of primary Sjögren syndrome (pSS). In this study, Raman spectra of 60 serum samples were acquired from 30 patients with pSS and 30 healthy controls (HCs). The means and standard deviations of the raw spectra of patients with pSS and HCs were calculated. Spectral features were assigned based on the literature. Principal component analysis (PCA) was used to extract the spectral features. Then, a particle swarm optimization (PSO)-support vector machine (SVM) was selected as the method of parameter optimization to rapidly classify patients with pSS and HCs. In this study, the SVM algorithm was used as the classification model, and the radial basis kernel function was selected as the kernel function. In addition, the PSO algorithm was used to establish a model for the parameter optimization method. The training set and test set were randomly divided at a ratio of 7:3. After PCA dimension reduction, the specificity, sensitivity and accuracy of the PSO-SVM model were obtained, and the results were 88.89%, 100% and 94.44%, respectively. This study showed that the combination of Raman spectroscopy and a support vector machine algorithm could be used as an effective pSS diagnosis method with broad application value. [ABSTRACT FROM AUTHOR]

As an emerging technology for artificial intelligence‐aided medical diagnosis, deep learning combined with Raman spectroscopy has great potential. The technology still has some problems in the actual medical diagnosis research process. The differences in spectrometers, experimental conditions, and experimental operations can result in non‐uniform and universally applicable data standards, which in turn lead to low data utilization. At the same time, it is still necessary to retrain the models when building diagnostic models for different diseases, which is time‐consuming and laborious. In this paper, a more complete transfer learning model for multiple types of serum Raman spectra is established for the first time, and a decision fusion strategy is applied to this diagnostic model. The Raman spectral data of serum from hepatitis B patients/control group, serum from abnormal thyroid function patients/control group, and serum from glioma patients/control group were selected as the source domains, and the Raman spectral data of tissue from hepatitis C patients/control group, serum from esophageal cancer patients/control group, and tissue from cervical cancer and cervical inflammation (patients/control) group were selected as the target domains. Three deep neural network models, ResNet, GoogLeNet, and CNN‐LSTM were trained in the source domain data for disease diagnosis, and the trained models were transfer to the target domain. The model is fine‐tuned by freezing different layers and then combined with logistic regression algorithms to construct a decision fusion model, which further improves the model effect. The results show that the proposed method can effectively improve the accuracy of transfer learning models. At the same time, this experiment extends the application of transfer learning in Raman spectroscopy and demonstrates that unrelated and scale‐different Raman datasets are still intrinsically connected, which also lays the foundation for us to build more stable and data‐inclusive spectral transfer learning fusion models in the future. [ABSTRACT FROM AUTHOR]

Membranous nephropathy is the main cause of nephrotic syndrome, which has an insidious onset and may progress to end-stage renal disease with a high mortality rate, such as renal failure and uremia. At present, the diagnosis of membranous nephropathy mainly relies on the clinical manifestations of patients and pathological examination of kidney biopsy, which are expensive, time-consuming, and have certain chance and other disadvantages. Therefore, there is an urgent need to find a rapid, accurate and non-invasive diagnostic technique for the diagnosis of membranous nephropathy. In this study, Raman spectra of serum and urine were combined with deep learning methods to diagnose membranous nephropathy. After baseline correction and smoothing of the data, Gaussian white noise of different decibels was added to the training set for data amplification, and the amplified data were imported into ResNet, AlexNet and GoogleNet models to obtain the evaluation results of the models for membranous nephropathy. The experimental results showed that the three deep learning models achieved an accuracy of 1 for the classification of serum data of patients with membranous nephropathy and control group, and the discrimination of urine data was above 0.85, among which AlexNet was the best classification model for both samples. The above experimental results illustrate the great potential of serum- and urine-based Raman spectroscopy combined with deep learning methods for rapid and accurate identification of patients with membranous nephropathy. [ABSTRACT FROM AUTHOR]

[Display omitted] • A novel spectral fusion diagnosis technology is proposed. • A new and efficient DBRAN fusion network is designed to process spectral data. • DBRAN fusion technology has achieved 100% accuracy, sensitivity and specificity in SLE disease classification, achieving significant performance improvement compared to single-modality (ORS or SERS) methods. As artificial intelligence technology gains widespread adoption in biomedicine, the exploration of integrating biofluidic Raman spectroscopy for enhanced disease diagnosis opens up new prospects for the practical application of Raman spectroscopy in clinical settings. However, for systemic lupus erythematosus (SLE), origin Raman spectral data (ORS) have relatively weak signals, making it challenging to obtain ideal classification results. Although the surface enhancement technique can enhance the scattering signal of Raman spectroscopic data, the sensitivity of the SERS substrate to airborne impurities and the inhomogeneous distribution of hotspots degrade part of the signal. To fully utilize both kinds of data, this paper proposes a two-branch residual-attention network (DBRAN) fusion technique, which allows the ORS to complement the degraded portion and thus improve the model's classification accuracy. The features are extracted using the residual module, which retains the original features while extracting the deep features. At the same time, the study incorporates the attention module in both the upper and lower branches to handle the weight allocation of the two modal features more efficiently. The experimental results demonstrate that both the low-level fusion method and the intermediate-level fusion method can significantly improve the diagnostic accuracy of SLE disease classification compared with a single modality, in which the intermediate-level fusion of DBRAN achieves 100% classification accuracy, sensitivity, and specificity. The accuracy is improved by 10% and 7% compared with the ORS unimodal and the SERS unimodal modalities, respectively. The experiment, by fusing the multimodal spectral, realized rapid diagnosis of SLE disease by fusing multimodal spectral data, which provides a reference idea in the field of Raman spectroscopy and can be further promoted to clinical practical applications in the future. [ABSTRACT FROM AUTHOR]

After collecting Raman spectral data of 12 compounds from commonly used Uyghur medicinal materials, the mixture data are synthesized using computer simulation methods. An accurate identification method for Uyghur medicine components based on Raman spectroscopy and multi-label deep learning is constructed, and the performance of the model Mix2Com is evaluated using several metrics such as accuracy, precision, recall value, and F1 score. [Display omitted] • Existing methods identifying Uyghur medicinal components aren't fit for large samples. • A new method based on Raman spectroscopy and multi-label learning is proposed. • Macro-parallel identification of components is achieved. • The model Mix2Com is constructed with LSTM and Attention introduced. • Mix2Com achieves 90.76% accuracy, better than XGBoost and DeepRaman. Uyghur medicine is one of the four major ethnic medicines in China and is a component of traditional Chinese medicine. The intrinsic quality of Uyghur medicinal materials will directly affect the clinical efficacy of Uyghur medicinal preparations. However, in recent years, problems such as adulteration of Uyghur medicinal materials and foreign bodies with the same name still exist, so it is necessary to strengthen the quality control of Uyghur medicines to guarantee Uyghur medicinal efficacy. Identifying the components of Uyghur medicines can clarify the types of medicinal materials used, is a crucial step to realizing the quality control of Uyghur medicines, and is also an important step in screening the effective components of Uyghur medicines. Currently, the method of identifying the components of Uyghur medicines relies on manual detection, which has the problems of high toxicity of the unfolding agent, poor stability, high cost, low efficiency, etc. Therefore, this paper proposes a method based on Raman spectroscopy and multi-label deep learning model to construct a model Mix2Com for accurate identification of Uyghur medicine components. The experiments use computer-simulated mixtures as the dataset, introduce the Long Short-Term Memory Model (LSTM) and Attention mechanism to encode the Raman spectral data, use multiple parallel networks for decoding, and ultimately realize the macro parallel prediction of medicine components. The results show that the model is trained to achieve 90.76% accuracy, 99.41% precision, 95.42% recall value and 97.37% F1 score. Compared to the traditional XGBoost model, the method proposed in the experiment improves the accuracy by 49% and the recall value by 18%; compared with the DeepRaman model, the accuracy is improved by 9% and the recall value is improved by 14%. The method proposed in this paper provides a new solution for the accurate identification of Uyghur medicinal components. It helps to improve the quality standard of Uyghur medicinal materials, advance the research on screening of effective chemical components of Uyghur medicines and their action mechanisms, and then promote the modernization and development of Uyghur medicine. [ABSTRACT FROM AUTHOR]

Screening and diagnosis of early‐stage lung cancer is low worldwide, and when lung cancer progresses to late stage, it will greatly affect the treatment and survival time of patients. Therefore, the development of an affordable and accurate diagnostic technique is critical for lung cancer patients. In this study, we analyzed and verified the changes of substance composition in the serum of early‐stage lung cancer patients and proposed an improved ResNeXt model to achieve accurate classification of serum Raman spectra of lung cancer. The robustness of the model is improved by adding different decibels of Gaussian white noise and spectral offset to augment the data, trying to process the augmented data with various pre‐processing methods, and inputting the pre‐processed data into the improved ResNeXt model, the improved ResNeXt model still shows the optimal performance after sufficient comparison experiments with three other more advanced deep learning models. The experimental results show that the improved ResNeXt model is very suitable for the classification of early‐stage lung cancer serum Raman spectra and may provide a reference for future early screening studies of other cancers. [ABSTRACT FROM AUTHOR]

Medical diagnosis technology based on convolutional neural networks (CNNs) has achieved good performance. In this study, we collected serum samples from 38 glioma patients and 45 healthy controls and used partial least squares (PLS) analysis to reduce the dimension of the data. Different levels of noise were added to the reduced data onto data augmentation, and the AlexNet, ResNet, and GoogLeNet fine‐tuning models were applied for classification. To evaluate the performance of the models, we used five‐fold cross‐validation. The accuracy rates of the AlexNet, ResNet, and GoogLeNet fine‐tuning models were 98.50%, 98.24%, and 99.50%, respectively. The model with the best classification effect was GoogLeNet. The specificity and sensitivity of this model were 98.98% and 98.48%, respectively. In addition, the area under the receiver operating characteristic (ROC) curve (AUC) of the established diagnostic model was 0.9949. The results showed that the combination of serum Raman spectroscopy and the PLS‐Gaussian‐GoogLeNet model achieved a good diagnostic effect for glioma. This method has high clinical application value and is worthy of further popularization. [ABSTRACT FROM AUTHOR]

The aim of the study is to evaluate the efficacy of the combination of Raman spectroscopy with feature engineering and machine learning algorithms for detecting glioma patients. In this study, we used Raman spectroscopy technology to collect serum spectra of glioma patients and healthy people and used feature engineering-based classification models for prediction. First, to reduce the dimensionality of the data, we used two feature extraction algorithms which are partial least squares (PLS) and principal component analysis (PCA). Then, the principal components were selected using the feature selection methods of four correlation indexes, namely, Relief-F (RF), the Pearson correlation coefficient (PCC), the F-score (FS) and term variance (TV). Finally, back-propagation neural network (BP), linear discriminant analysis (LDA) and support vector machine (SVM) classification models were established. To improve the reliability of the model, we used a fivefold cross validation to measure the prediction performance between different models. In this experiment, 33 classification models were established. Integrating 4 classification criteria, PLS-Relief-F-BP, PLS-F-Score-BP, PLS-LDA and PLS-Relief-F-SVM had better effects, and their accuracy rates reached 97.58%, 96.33%, 97.87% and 96.19%, respectively. The experimental results show that feature engineering can select more representative features, reduce computational time complexity and simplify the model. The classification model established in this experiment can not only increase the robustness of the model and shorten the discrimination time but also realize the rapid, stable and accurate diagnosis of glioma patients, which has high clinical application value. [ABSTRACT FROM AUTHOR]

In recent years, liver cancer has caused great harm to human health. This study aims to use serum Raman spectroscopy combined with deep learning algorithms to classify liver cancer patients and control groups. For improving the robustness of models, we added equal proportions of Gaussian white noise of 5, 10, 15, 20, 25 dBW to enhance the data, and compared the results of convolutional neural networks and long short-term memory networks which show that the convolutional neural network combined with 10-fold data enhancement was better. The accuracy was 96.95%, indicating the great potential of this experimental model in liver cancer detection. [ABSTRACT FROM AUTHOR]

Researchers have established a classification model based on tear Raman spectroscopy combined with machine learning classification algorithms, which realizes rapid noninvasive classification of cerebral infarction and cerebral ischemia, which is of great significance for clinical medical diagnosis. Through spectral data analysis, it is found that there are differences in the content of tyrosine, phenylalanine, and carotenoids in the tears of patients with cerebral ischemia and patients with cerebral infarction. We try to establish a classification model for rapid noninvasive screening of cerebral infarction and cerebral ischemia through these differences. The experiment has four parts, including normalization, data enhancement, feature extraction, and data classification. The researchers combined three feature extraction methods with four machine classification models to build a total of 12 classification models. Integrating 8 classification criteria, the classification accuracy of all models is above 85%, especially PLS-PNN has achieved 100% accuracy and better running time. The experimental results show that tear Raman spectroscopy combined with machine learning classification model has a good effect on the screening of cerebral ischemia and cerebral infarction, which is conducive to the noninvasive and rapid clinical diagnosis of cerebrovascular diseases in the future. [ABSTRACT FROM AUTHOR]

In this experiment, we collected 45 samples of cervicitis, 29 samples of low‐grade squamous intraepithelial lesion (LSIL), 44 samples of high‐grade squamous intraepithelial lesion (HSIL), 39 samples of cervical squamous cell carcinoma, and 38 cases of cervical adenocarcinoma. After preprocessing of the Raman spectral data, partial least squares (PLS) was used to reduce the dimensionality, and then extreme gradient boosting (XGBoost) was used for feature selection to obtain the first 30‐dimensional features. The preprocessed Raman spectral data also used a fast Fourier transform (FFT) to obtain amplitude information, and then PLS and XGBoost were used to obtain the first 30‐dimensional features. Finally, K nearest neighbor (KNN), extreme learning machine (ELM), artificial bee colony support vector machine (ABC‐SVM), support vector machine optimized by the cuckoo search algorithm (CS‐SVM), particle swarm optimization coupled with support vector machine (PSO‐SVM), and the convolutional neural network combined with long‐ and short‐term memory (CNN‐LSTM) classification models were established. In the raw Raman spectral features experiments, the classification accuracies of KNN, ELM, ABC‐SVM, CS‐SVM, PSO‐SVM, and CNN‐LTSM were 60.76%, 65.81%, 76.21%, 77.66%, 73.50%, and 69.19%, respectively. In the feature fusion experiments, the classification accuracies were 60.91%, 67.84%, 77.64%, 78.49%, 75.54%, and 70.72%, respectively. The experimental results show that feature fusion can further improve model performance regardless of whether using linear classification models or nonlinear classification models. Therefore, it provides a new strategy for extracting features and screening multiple cervical pathological tissues in the future. [ABSTRACT FROM AUTHOR]

Diabetic kidney disease (DKD) is one of the most common kidney diseases worldwide. It is estimated that approximately 537 million adults worldwide have diabetes, and up to 30%–40% of diabetic patients are at risk of developing nephropathy. The pathogenesis of DKD is complex, and its onset is insidious. Currently, the clinical diagnosis of DKD primarily relies on the increase of urinary albumin and the decrease in glomerular filtration rate in diabetic patients. However, the excretion of urinary albumin is influenced by various factors, such as physical activity, infections, fever, and high blood glucose, making it challenging to achieve an objective and accurate diagnosis. Therefore, there is an urgent need to develop an efficient, fast, and low-cost auxiliary diagnostic technology for DKD. In this study, an improved Dual Branch Attention Network (DBAN) was developed to quickly identify DKD. Serum Raman spectroscopy samples were collected from 32 DKD patients and 32 healthy volunteers. The collected data were preprocessed using the adaptive iteratively reweighted penalized least squares (airPLS) algorithm, and the DBAN was used to classify the serum Raman spectroscopy data of DKD. The model consists of a dual branch structure that extracts features using Convolutional Neural Network (CNN) and bottleneck layer modules. The attention module allows the model to learn features specifically, and lateral connections are added between the dual branches to achieve multi-level and multi-scale fusion of shallow and deep features, as well as local and global features, improving the classification accuracy of the experiment. The results of the study showed that compared to traditional deep learning algorithms such as Artificial Neural Network (ANN), CNN, GoogleNet, ResNet, and AlexNet, our proposed DBAN classification model achieved 95.4% accuracy, 98.0% precision, 96.5% sensitivity, and 97.2% specificity, demonstrating the best classification performance. This is the best method for identifying DKD, and has important reference value for the diagnosis of DKD patients, as well as improving the accuracy of medical auxiliary diagnosis. [Display omitted] • A new dual branch attention network (DBAN) is proposed. • The dual branch combines different attention mechanisms to achieve multi-level and multi-scale fusion. • The DBAN model achieves excellent classification performance for diabetic kidney disease diagnosis. [ABSTRACT FROM AUTHOR]

[Display omitted] • Differences in the biochemical composition of five types of cervical tissue were analyzed. • A multi-layered attention mechanism feature fusion architecture (MAFA) is proposed. • MAFA can effectively improve the classification accuracy of deep learning models for tissue samples. • MAFA as a model building method can be easily combined with a variety of deep learning models. Cervical cancer ranks among the most prevalent forms of gynecological malignancies. Timely identification of cervical lesions and prompt intervention can effectively prevent the development of cervical cancer or enhance patients' chances of survival. In this study, we propose an innovative method based on Raman spectroscopy, i.e., a multi-level SENet attention mechanism feature fusion architecture (MAFA) for rapid diagnosis of cervical cancer and precancerous lesions. The convolution process of this architecture can extract features from shallow to deep layers, and the attention mechanism is added to achieve the fusion of features from different layers. The added attention mechanism can automatically determine the importance of each layer feature channel and assign weight values to that layer according to the importance of each layer to achieve the purpose of focusing the model on certain waveform features and improve the targeting of model learning. We collected Raman spectra of 212 cervical tissues containing cervical cancer and its precancerous lesions.The experimental results show that MAFA can effectively improve the diagnostic accuracy of VGGNet, GoogLeNet and ResNet models in the validation of Raman spectral data of cervical tissue. Among them, ResNet performed the best, with the highest average accuracy, precision, recall and F1-Score of 82.36%, 84.00%, 82.35% and 82.26%, respectively, when no feature fusion was performed. The evaluation metrics improved by 4.91%, 3.97%, 4.97%, and 5.06%, respectively, after using the MAFA; they also improved by 4.16%, 2.90%, 4.17%, and 4.32%, respectively, compared with the model that directly performs feature fusion without using the attention mechanism. Therefore, the MAFA proposed in this study is better than that of the neural network that directly fuses the features of each convolutional layer. The experimental results show that the performance of the MAFA proposed in this paper is significantly higher than that of traditional deep learning algorithms, indicating that the present architecture can effectively improve the diagnostic accuracy of deep learning networks for cervical cancer. [ABSTRACT FROM AUTHOR]

This study presents a rapid method to screen hepatitis B patients using serum Raman spectroscopy combined with long short-term memory neural network (LSTM). The serum samples taken from 435 hepatitis B patients and 699 non-hepatitis B people were measured in this experiment. Specific biomolecular changes in three groups of serum samples could be seen in the tentative assignment of Raman peaks. First, principal component analysis (PCA) was used for extracting key features of spectral data, which reduces the dimension of the multidimensional spectrum. Then, LSTM is used to train the spectral data. Finally, the full connection layer completes the classification of HBV. The diagnostic accuracy of the first LSTM model is 97.32%, and the value of AUC is 0.995. The results from the study demonstrate that the combination of serum Raman spectroscopy technique and LSTM provides an effective technical approach to the screening of hepatitis B. [ABSTRACT FROM AUTHOR]

This study aimed to screen for thyroid dysfunction using Raman spectroscopy combined with an improved support vector machine (SVM). In spectral analysis, in order to further improve the classification accuracy of the SVM algorithm model, a genetic particle swarm optimization algorithm based on partial least squares is proposed to optimize support vector machine (PLS-GAPSO-SVM). In order to evaluate the performance of the algorithm, five optimization algorithms are used: grid search-based SVM (Grid-SVM), particle swarm optimization algorithm-based SVM (PSO-SVM), genetic algorithm-based SVM (GA-SVM), artificial fish coupled uniform design algorithm-based SVM (AFUD-SVM), and simulated annealing particle swarm optimization algorithm-based SVM (SAPSO-SVM). In this experiment, serum samples from 95 patients with confirmed thyroid dysfunction and 90 serum samples from normal thyroid function were used for Raman spectroscopy. The experimental results show that the GAPSO-SVM algorithm has a high average diagnostic accuracy of 95.08% and has high sensitivity and specificity (91.67%, 97.96%). Compared with the traditional optimization algorithm, the algorithm has high diagnostic accuracy, short execution time, and good reliability. It can be seen that Raman spectroscopy combined with GAPSO-SVM diagnostic algorithm has enormous potential in noninvasive screening of thyroid dysfunction. [ABSTRACT FROM AUTHOR]

In recent years, Raman spectroscopy combined with deep learning techniques has been widely used in various fields such as medical, chemical, and geological. However, there is still room for optimization of deep learning techniques and model compression algorithms for processing Raman spectral data. To further optimize deep learning models applied to Raman spectroscopy, in this study time, accuracy, sensitivity, specificity and floating point operations numbers(FLOPs) are used as evaluation metrics to optimize the model, which is named RamanCompact(RamanCMP). The experimental data used in this research are selected from the RRUFF public dataset, which consists of 723 Raman spectroscopy data samples from 10 different mineral categories. In this paper, 1D-EfficientNet adapted to the spectral data as well as 1D-DRSN are proposed to improve the model classification accuracy. To achieve better classification accuracy while optimizing the time parameters, three model compression methods are designed: knowledge distillation using 1D-EfficientNet model as a teacher model to train convolutional neural networks(CNN), proposing a channel conversion method to optimize 1D-DRSN model, and using 1D-DRSN model as a feature extractor in combination with linear discriminant analysis(LDA) model for classification. Compared with the traditional LDA and CNN models, the accuracy of 1D-EfficientNet and 1D-DRSN is improved by more than 20%. The time of the distilled model is reduced by 9680.9s compared with the teacher model 1D-EfficientNet under the condition of losing 2.07% accuracy. The accuracy of the distilled model is improved by 20% compared to the CNN student model while keeping inference efficiency constant. The 1D-DRSN optimized with channel conversion method saves 60% inference time of the original 1D-DRSN model. Feature extraction reduces the inference time of 1D-DRSN model by 93% with 94.48% accuracy. This study innovatively combines lightweight models and model compression algorithms to improve the classification speed of deep learning models in the field of Raman spectroscopy, forming a complete set of analysis methods and laying the foundation for future research. [Display omitted] • Proposed RamanCMP, a Raman spectral classification acceleration method. • Proposed 1D-EfficientNet and 1D-DRSN for spectral data. • Proposed channel conversion method and 1D-DRSN feature extractor. • Testing RamanCMP using the RRUFF public dataset and the external validation dataset for Sjögren's syndrome. [ABSTRACT FROM AUTHOR]

Abstract This study presents a rapid and non-invasive method to screen high renin hypertension using serum Raman spectroscopy combined with different classification algorithms. The serum samples taken from 24 high renin hypertension patients and 22 non-high renin hypertension samples were measured in this experiment. Tentative assignments of the Raman peaks in the measured serum spectra suggested specific biomolecular changes between the groups. Principal component analysis (PCA) was first used for feature extraction and reduced the dimension of high-dimension spectral data. Then, support vector machine (SVM), linear discriminant analysis (LDA) and k-nearest neighbor (KNN) algorithms were employed to establish the discriminant diagnostic models. The accuracies of 93.5%, 93.5% and 89.1% were obtained from PCA-SVM, PCA-LDA and PCA-KNN models, respectively. The results from our study demonstrate that the serum Raman spectroscopy technique combined with multivariate statistical methods have great potential for the screening of high renin hypertension. This technique could be used to develop a portable, rapid, and non-invasive device for screening high renin hypertension. Graphical abstract Raman spectroscopy technique combined with multivariate statistical methods for the screening of high renin hypertension. Unlabelled Image Highlights • Raman spectroscopy has the potential to screen high renin hypertension patients. • The high accuracy of 93.5% is obtained from multivariate statistical method. • Our method is expected to develop a clinical diagnostic tool. [ABSTRACT FROM AUTHOR]

Raman spectroscopy has been widely used in discriminant analysis. In order to improve the accuracy of Raman spectroscopy discrimination, a model combining kernel principal component analysis (KPCA) and support vector machine (SVM) is proposed. Firstly, the Raman spectral discriminant data is collected, which is subjected to the fifth-order polynomial smoothing and vector normalization preprocessing to eliminate the influence of noise. Then, the collected unbalanced data is oversampled by the Synthetic Minority Over-sampling Technique algorithm, and the KPCA method is used to extract the features of the balanced data. The SVM discriminant model is established by selecting different kernel functions for the extracted features. In order to evaluate the performance of the KPCA-SVM discriminant model, it is compared with the PCA-SVM discriminant model under the same experimental conditions. The experimental results show that the KPCA-SVM discriminant model achieves a discriminative accuracy rate of 93.75%, which is better than that of the PCA-SVM discriminant model (87.5%). This study provides a new idea for improving the discrimination accuracy of Raman spectroscopy. [ABSTRACT FROM AUTHOR]

[Display omitted] • In this paper, a new spectral analysis method based on the spectral seal Minkowski distance is proposed. • This paper also presents an auxiliary alcohol dependence diagnosis technique based on Raman spectral structure distance feature and a convolutional neural network for the first time. • We evaluated the good performance of the proposed technique of the study. • The results demonstrate that this study provides a new scheme for auxiliary diagnosis and medical Raman spectroscopy analysis of alcohol dependence. Alcohol dependence is an addictive disease, which can lead to cancer, depression, and other problems. The current diagnostic methods of alcohol dependence are easily disturbed by patients' subjective consciousness, which could lead to wrong diagnoses. To solve this problem, this paper applies serum Raman spectra and a deep learning algorithm to the diagnosis of alcohol dependence and constructs an innovative diagnostic framework. We collected 228 serum samples for spectral analysis, including 119 alcohol-dependent patients and 109 controls. In this framework, spectra are firstly converted to two-dimensional spectrograms by clipping and stitching. Then, a calculating box extracts the distance features between spectral nonadjacent peaks on the spectrogram. Finally, the eigenvalues of each peak are used as pixels of artificial images and input into the ConvNeXt-T neural network model for classification. The sensitivity and specificity of the method are 81.82% and 100%. The results show that it is feasible to use Raman spectroscopy combined with a deep learning algorithm to diagnose alcohol dependence. Compared with directly using the ConvNeXt-T model, the distance feature performs better. [ABSTRACT FROM AUTHOR]

[Display omitted] • For the first time, Raman spectroscopy combined with deep learning is used for the classification and identification of red wine origin. • Corresponding biomarkers identified by Raman spectroscopy of red wine. • Classification of red wine origin by PCA analysis. • Comparative experiment of four deep learning models. • The combination of MCNN and Raman spectroscopy achieves the best results for the classification of red wine origin. In this study, we combined Raman spectroscopy with deep learning for the first time to establish an accurate, simple, and fast method to identify the origin of red wines. We collected Raman spectra from 200 red wine samples of the Cabernet Sauvignon variety from four different origins with a portable Raman spectrometer. The red wine samples, made in 2021, were from the same producer in China. Differences were found by analyzing the Raman spectra of red wine samples. These differences are mainly caused by ethanol, carboxylic acids, and polyphenols. After further analysis, for different origins, the different performances of these substances on the Raman spectrum are related to the climate and geographical conditions of the origin. The Raman spectra were analyzed by principal component analysis (PCA). The data with PCA dimensionality reduction were imported into an artificial neural network (ANN), multifeature fusion convolutional neural network (MCNN), GoogLeNet, and residual neural network (ResNet) to establish red wine origin identification models. The classification results of the model prove that climate, geography, and other conditions can provide support for the classification of red wine origin. The experiments showed that all four models performed well, among which MCNN performed the best with 93.2% classification accuracy, and the area under the curve (AUC) was 0.987. This study provides a new means to classify the origin of red wine and opens up new ideas for identifying origins in the food field. [ABSTRACT FROM AUTHOR]

[Display omitted] • Differences in the biochemical composition of seven types of tissues are analyzed. • A new one-dimensional convolutional hierarchical network (H-CNN) is proposed. • H-CNN classifies tissue samples in a hierarchy from coarse to fine. • H-CNN outperforms traditional classification models. Cervical cancer is one of the most common cancers with a long latent period and slow onset process. Early and accurate identification of the stage of cervical cancer can significantly improve the cure rate and patient survival time. In this study, we collected 699 Raman spectral data of tissue sections from 233 different patients. We analyzed and compared the characteristics and differences of the mean Raman spectra of the seven tissues and pointed out the main differences in the biochemical composition of the seven tissues. In this study, 1D hierarchical convolutional neural network (H-CNN) is proposed by integrating the prior knowledge of hierarchical classification relations with the research of deep learning in Raman spectroscopy. H-CNN is based on CNN and is added with three network branches. Hierarchical classification is performed from coarse to fine for tissue samples of cervicitis, Low-grade Squamous Cell Carcinoma, High-grade Squamous Cell Carcinoma, Well Differentiated Squamous Cell Carcinoma, Moderately Differentiated Squamous Cell Carcinoma, Poorly Differentiated Squamous Cell Carcinoma and cervical adenocarcinoma. To evaluate the recognition performance of H-CNN, we compared it with traditional methods such as Bayesian classifier (NB), decision tree classifier (DT), support vector machine classifier (SVM) and CNN. The experimental results show that H-CNN can accurately identify different classes of tissue sections and has apparent advantages in several aspects such as recognition accuracy, stability and sensitivity compared with the other four traditional recognition methods. The classification Macro-Accuracy of H-CNN can reach 94.91%, Macro-Recall can reach 95.31%, Macro-F1 can reach 95.23%, and Macro-AUC can reach 97.35%. The hierarchical classification method proposed in this study can diagnose patients more accurately. This could lay the foundation for further research on Raman spectroscopy as a clinical diagnostic method for cervical cancer. [ABSTRACT FROM AUTHOR]

Raman spectroscopy combined with artificial intelligence (AI) is widely used in medical diagnostic research and has great application value. However, there are still problems in the research process, such as the low prevalence of some diseases and difficulties in obtaining research samples, which will easily lead to data imbalance in medical Raman spectroscopy research. For AI classification and diagnosis algorithms, when the data imbalance problem is not addressed, majority class samples are selected, and the importance of minority class samples is ignored, reducing the accuracy of disease identification. Based on the above problems, this paper proposes a hybrid sampling technique of Raman-Gaussian distributed oversampling fused with random undersampling (R-GDORUS) to solve the data imbalance problem in medical Raman spectroscopy. The density and distance information carried by the minority samples are used to obtain the selection probability of the minority samples, determine the anchor samples from the minority samples, and generate a new minority sample in the form of a Gaussian distribution. Finally, a random undersampling strategy is used to remove some of the majority class spectral samples. This technique and five other mainstream methods for handling imbalanced data are applied to three major types of imbalanced medical Raman spectroscopy datasets: malignant tumors, class B infectious diseases and autoimmune diseases, and the performance of the technique is evaluated using the AUC and G-mean values. The results demonstrate that the proposed technique can be used to effectively reduce the impact of impaired model performance caused by spectral data imbalance and has good application prospects. • This paper presents a hybrid sampling technique R-GDORUS for the first time. • We evaluated the good performance of the proposed technique of the study. • The results show that the method proposed provides a new solution to the Raman data imbalance in medical studies. [ABSTRACT FROM AUTHOR]

[Display omitted] • Triple-negative breast cancer and HER2-positive breast cancer are highly malignant, and early diagnosis is helpful to improve the chances of cure. • Serum Raman spectroscopy combined with deep learning algorithms to diagnose triple-negative breast cancer and HER2-positive breast cancer and achieved great diagnostic results. • The proposed method is helpful for early breast cancer diagnosis and can provide a reference for breast cancer patients to make a personalised diagnosis. Breast cancer is common in women, and its number of patients ranks first among female malignant tumors. Breast cancer is highly heterogeneous, and different types of breast cancer have different biological behaviors and prognoses. Therefore, identifying the different types of breast cancer is of great help in formulating individualized treatment plans. Based on serum Raman spectroscopy and deep learning algorithms, we propose a fast and low-cost diagnosis method for screening triple-negative breast cancer, human epidermal growth factor receptor 2 (HER2)-positive breast cancer, and healthy controls. We collected 75 serum samples in this study, including 23 triple-negative breast cancers, 22 HER2-positive breast cancers, and 30 healthy controls. Using the preprocessed Raman spectra as the input of deep learning, three deep learning models, neural network language model (NNLM), bidirectional long-short-term memory network (BiLSTM), and convolutional neural network (CNN), were established, and the accuracy rates of the three models were 87.78%, 90.37%, and 91.11%, respectively. The experimental results demonstrate the feasibility of serum Raman spectroscopy combined with deep learning algorithms to diagnose breast cancer, which can be used as an effective auxiliary diagnosis method for breast cancer. [ABSTRACT FROM AUTHOR]

[Display omitted] • A novel spectral fusion diagnosis technology is proposed. • A new and efficient MFCNN network is designed to process spectral data. • This study explored the potential of spectral fusion technology for the diagnosis of multiple cancers in this complex clinical environment. According to the limited molecular information reflected by single spectroscopy, and the complementarity of FTIR spectroscopy and Raman spectroscopy, we propose a novel diagnostic technology combining multispectral fusion and deep learning. We used serum samples from 45 healthy controls, 44 non-small cell lung cancer (NSCLC), 38 glioma and 37 esophageal cancer patients, and the Raman spectra and FTIR spectra were collected respectively. Then we performed low-level fusion and feature fusion on the spectral, and used SVM, Convolutional Neural Network-Long-Short Term Memory (CNN-LSTM) and the multi-scale convolutional fusion neural network (MFCNN). The accuracy of low-level fusion and feature fusion models are improved by about 10% compared with single spectral models. [ABSTRACT FROM AUTHOR]

This research innovatively combines Fourier transform infrared spectra, Raman spectra and its first-derivative spectra to develop a rapid diagnosis method for renal cell carcinoma (RCC). The spectral features are extracted by principal component analysis (PCA), and then the normalized fused data are used as input to the improved AlexNet model and MCNN model to train the models. Among them, the AlexNet model performed better, with a classification accuracy rate of 93%. [Display omitted] • It is proposed for the first time to combine FT-IR, Raman spectra and their first-derivative spectra to develop a rapid diagnosis method for renal cell carcinoma. • The accuracy of the multi-spectral fusion method (93%) is greatly improved compared with that of single infrared spectroscopy (74%), signal Raman spectroscopy (75%) and the fusion of infrared spectroscopy and Raman spectroscopy (79%). • It shows that multi-spectral fusion technology combined with deep learning algorithms has great potential in the diagnosis of RCC. This research innovatively combines FT-IR, Raman spectroscopy and their first-derivative spectroscopy to develop a rapid diagnosis method for renal cell carcinoma (RCC). After measuring the Raman spectra and FT-IR spectra of 45 cases of control subjects and 28 cases of RCC, the first derivative of the infrared spectra and the Raman spectra were calculated respectively. Principal component analysis (PCA) was used to extract the features of the infrared spectra, first-derivative infrared spectra, Raman spectra and first-derivative Raman spectra. Then the four feature matrices were merged as fused spectral feature matrices. The fused matrices were used as the input of AlexNet and MCNN. The fused spectral feature matrices were used as the input of AlexNet and MCNN. The adjusted AlexNet model performed better, and the classification accuracy of the fused spectral data is 93%. Compared with the classification results of infrared spectra (74%), Raman spectra (75%) and the fusion of infrared and Raman spectra (79%) combined with the adjusted AlexNet model, the classification result of the fusion of infrared spectra, Raman spectra and their first-derivative was significantly improved. The experimental results show that infrared spectroscopy, Raman spectroscopy and their first-derivative fusion technology combined with deep learning algorithms has great potential in the diagnosis of RCC. [ABSTRACT FROM AUTHOR]

In this research, PLSR, GA-SVR and ELM models combined with sheep serum Raman spectroscopy were used for quantitative detection, the results of PLSR determination are better (a). In order to further improve the performance of the model, the SPA is used for band selection (c). The optimized quantitative model result (b) of DOSC-SPA-PLSR-MN performs well. • It is proposed for the first time to use sheep serum Raman spectroscopy combined with a quantitative model to determine the concentration (/enzyme activity) of seven indicators. • It is the first time to deviation standardization of the model evaluation index for many kinds of substances. • The comparison results of SPA band selection show that there is a correlation among the indicators in the serum. • The optimized DOSC-SPA-PLSR-DS quantitative model shows that the average R p 2 of the seven indicators is 0.9834. Hepatic fascioliasis, ketosis of pregnancy, toxemia of pregnancy and other common sheep diseases will directly affect the concentration (/enzymatic activity) of seven indicators, such as cortisol and high-density lipoprotein cholesterol (HDL-C) in sheep serum. Whether the concentrations (/enzymatic activity) of these indicators can be detected quickly will directly affect the prevention of sheep diseases and the targeted adjustment of breeding methods, thereby affecting the economic benefits of sheep breeding. In this research, we established partial least square regression (PLSR), support vector regression based on genetic algorithm optimization (GA-SVR) and extreme learning machine (ELM) models. Due to the large differences in the content of different substances, it is difficult to directly use the RMSE to evaluate the quantitative effect of the model. This study is the first to propose conducting deviation standardization (DS) for the determination results of various substances. To further improve the performance of the model, we use the successive projections algorithm (SPA) to optimize feature extraction and combine it with the better-performing PLSR model for training. The results show that the optimized DOSC-SPA-PLSR-DS quantitative model has better determination results for 101 sheep serum samples. The average RMSE p* of the concentration of the six substances decreased from 0.0408 to 0.0387, the R p 2 increased from 0.9758 to 0.9846, and the running time was reduced from 0.1659 to 0.0008 s. And the determination performance of lipase (LPS) enzymatic activity has also been improved. The results of this research show that sheep serum Raman spectroscopy combined with DOSC-SPA-PLSR-DS optimization can efficiently monitor the concentration (/enzyme activity) of seven indicators in real time and provide a new strategy for future intelligent supervision of animal husbandry. [ABSTRACT FROM AUTHOR]

In this study, the Raman spectra of serum samples from patients with echinococcosis and healthy controls were recorded by a portable spectrometer under 532 nm excitation. Due to the resonance Raman effect, the two Raman peaks (at 1154 and 1515 cm-1) belonging to β carotene were significantly enhanced, resulting in relatively weak signal values for other peaks. In order to amplify the weak Raman peaks and improve the capability of Raman spectroscopy for the differentiation of healthy and echinococcosis infected serum samples, a logarithmic transformation preprocessing method was used. Furthermore, principal component analysis (PCA), combined with linear discriminant analysis (LDA), were used to distinguish echinococcosis patients from healthy volunteers. The results show that the diagnostic models formed by the log-transformed spectral data sets perform better in the balance of sensitivity and specificity. The accuracy, sensitivity, and specificity of the diagnostic model based on the spectral data set of optimal scale transformation were 98.00%, 98.10%, and 98.00%, respectively, which were slightly superior to the untransformed spectral data set. Our findings suggest that serum log-transformed Raman spectroscopy combined with the PCA-LDA method has great potential for improving the detection of echinococcosis. [ABSTRACT FROM AUTHOR]

In this study, the Raman spectra of serum samples from patients with echinococcosis and healthy controls were recorded by a portable spectrometer under 532 nm excitation. Due to the resonance Raman effect, the two Raman peaks (at 1154 and 1515 cm-1) belonging to β carotene were significantly enhanced, resulting in relatively weak signal values for other peaks. In order to amplify the weak Raman peaks and improve the capability of Raman spectroscopy for the differentiation of healthy and echinococcosis infected serum samples, a logarithmic transformation preprocessing method was used. Furthermore, principal component analysis (PCA), combined with linear discriminant analysis (LDA), were used to distinguish echinococcosis patients from healthy volunteers. The results show that the diagnostic models formed by the log-transformed spectral data sets perform better in the balance of sensitivity and specificity. The accuracy, sensitivity, and specificity of the diagnostic model based on the spectral data set of optimal scale transformation were 98.00%, 98.10%, and 98.00%, respectively, which were slightly superior to the untransformed spectral data set. Our findings suggest that serum log-transformed Raman spectroscopy combined with the PCA-LDA method has great potential for improving the detection of echinococcosis. [ABSTRACT FROM AUTHOR]

This paper proposes for the first time to establish grid search-support vector regression (GS-SVR), random forest (RF) and partial least squares regression (PLSR) models based on the measured Raman spectral data of sheep serum to find a method that can quickly detect glucose (GLU) concentration. The results show that all performance indexes of PLSR model are optimal, root mean square error(R M S E t r a i n) and determination coefficient(R t r a i n 2) of training set are 0.1034 and 0.9726 respectively, and root mean square error(R M S E t e s t) and determination coefficient(R t e s t 2) of test set are 0.0691 and 0.9632 respectively. Studies have shown that the PLSR model has good predictive performance for the detection of glucose in sheep serum. [ABSTRACT FROM AUTHOR]

In this study, we propose a method of tear Raman spectroscopy combined with support vector machine (SVM) to accurately, quickly and economically diagnose cerebral infarction. Tear samples from 15 patients with cerebral infarction and 47 healthy volunteers were assessed, and their Raman spectra were compared. Principal component analysis (PCA) was used for feature extraction. Then, support vector machine (SVM) was used to establish an efficient diagnostic model. To improve the accuracy of the SVM diagnostic model, we used the linear, polynomial, RBF and sigmoid kernel functions of SVM for discrimination and evaluation, and the best results were obtained when the RBF kernel was used, with an accuracy of 96.77 %, a specificity of 100 %, and a sensitivity of 86.67 %. Our results show that the combination of tear Raman spectroscopy and the SVM algorithm has great potential for screening cerebral infarction. This technology is expected to aid in the development of a portable, economic and high accuracy early-assisted screening device for cerebral infarction. [ABSTRACT FROM AUTHOR]

In this study, serum samples from 199 patients with thyroid dysfunction and 183 people with normal thyroid function were collected by Raman spectroscopy, and the data were dimensions-reduced by PCA. The reduced data were input into a multi-feature fusion convolutional neural network (MCNN), the improved AlexNet, VGGNet, GoogLeNet and ResNet, Support Vector Machine (SVM) and Decision Tree (DT) for classification, and the results of the seven classification algorithms were compared. Their classification accuracy are 94.01 %, 91.91 %, 90.34 %, 93.46 %, 92.42 %, 82.78 % and 80.89 %, respectively. The results of this study indicate that the combination of serum Raman spectra and MCNN has a good diagnostic effect for identifying thyroid dysfunction, and it is feasible to improve the classic deep learning models for Raman spectrum classification. [ABSTRACT FROM AUTHOR]

In this study, a rapid, noninvasive and efficient screening method for human papillomavirus (HPV) was proposed by combining cervical secretion Raman spectral data with a pattern recognition algorithm. We measured the cervical secretion Raman spectra of 58 HPV positive and 196 HPV negative samples. The preliminary assignment of the Raman peaks in the measured cervical secretion spectra indicated changes in specific biomolecules between the groups, and we analyzed them according to their differences. The specificity of the airPLS-PLS-GA-SVM model was 100 %, the sensitivity was 94.4 %, and the total accuracy was 98.6 %. The results show that the combination of cervical secretion Raman spectroscopy and multivariate statistical analysis has great potential in the early screening of HPV and the samples of low-level lesions and high-level lesions were preliminarily classified effectively, providing a basis for the design of computer-aided diagnosis systems in the future. [ABSTRACT FROM AUTHOR]

The survival time of patients with pancreatic tumors is closely related to the type of tumor. To diagnose patients with malignant pancreatic tumors as early as possible in order to improve patient survival, we proposed a new method for the assisted diagnosis of pancreatic tumors. First, we used partial least squares (PLS) to extract the Raman spectrum information of patients with pancreatic tumors. The study extracted ten eigenvalues experimentally. The cumulative variance of the first six PLS components reached 97.045 %, and then the feature information extracted by PLS was classified. This experiment used three classification algorithms: linear discriminant analysis (LDA), support vector machine (SVM), and k-nearest neighbor (KNN). Among them, the cubic kernel of SVM achieved the best classification effect, and its classification accuracy reached 96.4 %. This is the first time that serum Raman spectroscopy has been used to distinguish patients with benign and malignant tumors of the pancreas. The experimental results show that serum Raman spectroscopy may become a new auxiliary method for the clinical diagnosis of pancreatic cancer. [ABSTRACT FROM AUTHOR]

This study presents a method combining serum Raman spectroscopy with a multiscale convolution independent circulation neural network (MSCIR) for hepatitis B virus (HBV) diagnosis. It simplifies the steps of serum Raman spectroscopy data preprocessing and improves the accuracy of hepatitis B diagnosis. Serum samples were obtained from 499 healthy people and 435 HBV patients. First, feature extraction of serum Raman spectroscopy data through principal component analysis (PCA) was performed to reduce the dimensionality of spectral data. Then, the linear discriminant analysis (LDA), k-nearest neighbor (KNN), support vector machine (SVM), random forest (RF), artificial neural network (ANN) and MSCIR algorithms were employed to establish the discriminant diagnostic models. Their accuracies are 80.77%, 77.69%, 89.23%, 86.92%, 91.53% and 96.15%, respectively. The results show that the MSCIR prediction accuracy is higher than that of the five traditional algorithms, and the fitness is stable. Therefore, the MSCIR algorithm can effectively diagnose hepatitis B patients. [ABSTRACT FROM AUTHOR]

Chronic renal failure (CRF) is a symptom that is caused by kidney damage that deteriorates to the end stage. If not treated in time, CRF will worse into uraemia, which greatly reduces the lifespan of the patient. However, current screening strategies, including routine blood and medical image investigations, have poor sensitivity. Therefore, exploring new and efficient diagnostic methods such as urine spectroscopy for CRF is of great significance. In this study, we first explored Raman spectroscopy to classify urine from CRF patients and control subjects with normal renal function. A total of 48 samples from CRF patients and 44 samples from control subjects were accrued. The spectra revealed relatively lower hydroxybutyrate and higher alanine, creatinine and porphyrin in CRF. Subsequent principal component analysis (PCA) was first used for feature extraction. Then, back propagation (BP), grid search support vector machine (GS-SVM), genetic algorithms based on support vector machine (GA-SVM), discriminant analysis (DA) and particle swarm optimization support vector machine (PSO-SVM) algorithms were employed to establish discriminant diagnostic models; the diagnostic accuracy of each of the five classifiers was 70.77 %, 84.62 %, 80.77 %, 65.20 % and 74.62 %, respectively, for control subjects and CRF patients. The results show the potential of Raman spectroscopy in rapid screening of CRF urine samples. Based on the limitations of current routine diagnostic methods, urine Raman spectroscopy may be a replaceable method for the clinical diagnosis of CRF with the prospective validation of more samples. [ABSTRACT FROM AUTHOR]