Savitzky-Golay Algorithm Research Articles

Research on the Method of Imperfect Wheat Grain Recognition Utilizing Hyperspectral Imaging Technology

As the primary grain crop in China, wheat holds a significant position in the country’s agricultural production, circulation, consumption, and various other aspects. However, the presence of imperfect grains has greatly impacted wheat quality and, subsequently, food security. In order to detect perfect wheat grains and six types of imperfect grains, a method for the fast and non-destructive identification of imperfect wheat grains using hyperspectral images was proposed. The main contents and results are as follows: (1) We collected wheat grain hyperspectral data. Seven types of wheat grain samples, each containing 300 grains, were prepared to construct a hyperspectral imaging system for imperfect wheat grains, and visible near-infrared hyperspectral data from 2100 wheat grains were collected. The Savitzky–Golay algorithm was used to analyze the hyperspectral images of wheat grains, selecting 261 dimensional effective hyperspectral datapoints within the range of 420.61–980.43 nm. (2) The Successive Projections Algorithm was used to reduce the dimensions of the 261 dimensional hyperspectral datapoints, selecting 33 dimensional hyperspectral datapoints. Principal Component Analysis was used to extract the optimal spectral wavelengths, specifically selecting hyperspectral images at 647.57 nm, 591.78 nm, and 568.36 nm to establish the dataset. (3) Particle Swarm Optimization was used to optimize the Support Vector Machines model, Convolutional Neural Network model, and MobileNet V2 model, which were established to recognize seven types of wheat grains. The comprehensive recognition rates were 93.71%, 95.14%, and 97.71%, respectively. The results indicate that a larger model with more parameters may not necessarily yield better performance. The research shows that the MobileNet V2 network model exhibits superior recognition efficiency, and the integration of hyperspectral image technology with the classification model can accurately identify imperfect wheat grains.

Resveratrol enhances the antiliver cancer effect of cisplatin by targeting the cell membrane protein PLA2.

In this study, we aimed to explore the mechanism by which resveratrol promotes cisplatin-induced death of HepG2 cells and to provide a potential strategy for resveratrol in the treatment of cancer. HepG2 cells were exposed to a range of drug concentrations for 24 h: resveratrol (2.5 μg/mL [10.95 μM], 5 μg/mL [21.91 μM], 10 μg/mL [43.81 μM], 20 μg/mL [87.62 μM], 40 μg/mL [175.25 μM], and 80 μg/mL [350.50 μM]), cisplatin (0.625 μg/mL [2.08 μM], 1.25 μg/mL [4.17 μM], 2.5 μg/mL [8.33 μM], 4.5 μg/mL [15.00 μM], and 10 μg/mL [33.33 μM]), 24 μg/mL (105.15 μM) resveratrol + 9 μg/mL (30.00 μM) cisplatin, and 12 μg/mL (52.57 μM) resveratrol + 4.5 μg/mL (15.00 μM) cisplatin. The interaction of two drugs was evaluated by coefficient of drug interaction (CDI), which was based on the Pharmacological Additivity model. The MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay was used to detect the effect of different concentrations of drugs on cell viability, whiletranscriptome sequencing was used to identify pathways associated with higher gene enrichment. Synchrotron radiation FTIR microspectroscopy experiments and data analysis were conducted to obtain detailed spectral information. The second-derivative spectra were calculated using the Savitzky-Golay algorithm. Single-cell infrared spectral absorption matrices were constructed to analyze the spectral characteristics of individual cells. The Euclidean distance between cells was calculated to assess their spectral similarity. The cell-to-cell Euclidean distance was computed to evaluate the spatial relationships between cells. The target protein of resveratrol was verified by performing a Western blot analysis. After 24 h of treatment with resveratrol, HepG2 cell growth was inhibited in a dose-dependent manner. Resveratrol promotes cisplatin-induced HepG2 cell death through membrane-related pathways. It also significantly changes the membrane components of HepG2 cells. Additionally, resveratrol changes the morphology of the HepG2 cell membrane by decreasing the expression of PLA2G2. Resveratrol changes the morphology of the HepG2 cell membrane by decreasing the expression of PLA2G2 and promotes cisplatin-induced HepG2 cell death. The combination of cisplatin and resveratrol can play a synergistic therapeutic effect on HepG2 cells.

Ion beam stability prediction of ECR ion source based on TCN-DTW network

The Electron Cyclotron Resonance (ECR) ion source is an irreplaceable apparatus for producing high-intensity, highly charged heavy ion beams, representing a critical component for heavy ion accelerators. The operation of the ECR ion source is inherently influenced by various factors, leading to fluctuations in beam intensity. Such instability not only diminishes the efficacy of accelerator operations but also introduces distortions in terminal experimental data. Addressing these challenges, this study proposes the application of a Temporal Convolutional Network (TCN) based on a Dynamic Time Warping (DTW) loss function (TCN-DTW) for predicting the stability of the ion beams. Prior to constructing the prediction network, raw data undergoes preprocessing through an Interquartile Range (IQR) anomaly detection mechanism and the Savitzky-Golay (SG) filtering algorithm with an adaptive window. Experimental results demonstrate a substantial enhancement in prediction performance when employing the TCN network with the DTW loss function compared to traditional alternatives. This approach facilitates effective forecasting of the ion source beam current trend, offering a basis for the control and correction of long-term stability. Consequently, it provides valuable insights for optimizing the ECR ion source and enhancing overall accelerator operational performance.

Multidimensional information fusion and analysis of ultraviolet absorption spectroscopy: Simultaneous detection method for copper and zinc ion concentrations in seawater

Heavy metal pollution poses a significant threat to marine ecological environments and serves as a crucial indicator for seawater quality assessment. The enrichment of heavy metals severely impacts marine ecosystem health and poses threats to human food safety, underscoring the critical importance of monitoring heavy metal levels in marine waters. In this paper, UV absorption spectroscopy was used to collect the spectra of seawater containing different concentrations of Cu and Zn ions, which overcomes the problems of traditional methods that require extensive pre-treatment and lossy samples. To address the characteristics of the data, the Savitzky-Golay (SG) filtering algorithm is used for preprocessing. The acquired spectra are then subjected to further data analysis using the proposed multidimensional information fusion method called SpectraNet Fusion Algorithm (SFA) for UV absorption spectroscopy. The local feature information extraction module and the whole domain information feature extraction module were respectively used to predict the regression of the data in parallel two-channel, and the output results were integrated to form a new feature map, which was further processed by the feature information fusion module, and finally output the prediction results corresponding to the spectra to realize the accurate detection of Cu and Zn heavy metal ions in seawater. The coefficient of determination (R2) of the model training set established by this method reaches 98.00 %, the root mean square error (RMSE) reaches 0.0389, and the mean absolute error (MAE) reaches 0.0243, combined with the experimental design, the detection range was 5 μg·L−1 to 90 μg·L−1,which realizes simultaneous and high-precision prediction of Cu and Zn heavy metal ions in seawater.

Measurement of Ammonia Leakage by TDLAS in Mid-Infrared Combined with an EMD-SG Filter Method

Ammonia (NH3) has been used as the main refrigerant in various types of cold storages, exceeding 90% of cooling applications. However, NH3 leakage occurs frequently due to damage of the refrigeration compressor, resulting in environmental pollution and intoxication accidents. Therefore, reliable and highly sensitive NH3 detection is a critical requirement in cold storage. In this paper, tunable diode laser absorption spectroscopy (TDLAS) is used to measure NH3 leakage, where a new denoising method combining empirical mode decomposition with the Savitzky-Golay smoothing algorithm (EMD-SG) is proposed to improve the signal-to-noise ratio (SNR) of absorbance signals. The results show that a 5.5 times improvement of SNR and 86% reduction of the linear fitting error in calibration is obtained; furthermore, this system achieves a detection limit of 0.2 ppm at an average time of 69 s by Allan variance analysis. Hence, it can be concluded that the EMD-SG algorithm can effectively improve the performance of TDLAS systems to meet the requirements for NH3 leakage detection in cold storages.

Noise Processing of Ocean Absolute Gravity Data Based on Savitzky-golay Filtering Algorithm

High-precision data on absolute gravity have a wide range of applications, including basic science, resource exploration, national defense infrastructure, seismic studies, and inertial navigation, among others. The atomic gravimeter, known for its high sampling rate, accuracy and long-term stability, is particularly well suited for high-precision measurements of absolute gravity in oceanic and deep-sea environments. However, the challenging conditions found in the ocean introduce significant noise into the gravity data obtained from atomic gravimeters. This noise adversely affects the accuracy of gravity measurements, which in turn affects the interpretation of geological structures. Consequently, the analysis and processing of noise in atomic gravimeter data has become a focal point in the research of high-precision absolute oceanic gravity dynamic measurements. In this study, we conducted absolute gravity measurement experiments using the shipboard atomic absolute gravity measurement system developed at Zhejiang University of Technology. To filter the collected gravity data, we used the Savitzky-Golay (SG) filtering algorithm. This algorithm not only preserves essential measurement information, but also provides superior accuracy compared to conventional filtering methods.

FTIR fingerprint — testing a new representation of the binary fingerprint based on FTIR spectra in the prediction of physicochemical properties

The paper deals with the development of a new method for the generation of binary fingerprints based on the Savitzky-Golay (SG) algorithm and first-order derivatives of FTIR spectra, which are then used to create prediction models for selected the physicochemical properties of chemical compounds. Models based on the FEDS (Functionally-Enhanced Derivative Spectroscopy) transformation and raw spectra were used as a reference to determine whether the use of the SG filter and first-order derivatives was worth to further develop. The FTIR spectra of 103 compounds with theoretically determined values of logP, logD and logS were studied. The Tanimoto coefficient and correlation coefficient were used to compare the fingerprints obtained, while the root mean square error (RMSE) was used to assess the quality of the prediction models. Based on the results, it was found that the use of the SG filter and derivatives had a positive effect on the quality of the prediction models for logP and logS, and a negative effect on the quality of the models for logD, compared to the models based on original spectra and FEDS transformation.

Non-Destructive Classification of Chrysanthemum Tea Using Near-Infrared Spectroscopy (NIRS) and Fuzzy Improved Pseudoinverse Linear Discriminant Analysis (FIPLDA)

Chrysanthemum tea is popular because of its health benefits, although different varieties vary in effectiveness and role. The rapid identification of the variety is critical for consumers and the tea industry. This study employed near-infrared (NIR) spectroscopy in combination with feature extraction methods to characterize chrysanthemum tea varieties. To improve the recognition accuracy, a fuzzy improved pseudoinverse linear discriminant analysis (FIPLDA) algorithm was employed to extract discriminant information from the NIR spectra of chrysanthemum tea that were preprocessed using the Savitzky-Golay (SG) algorithm. Next, the dimensionality of the data was reduced by principal component analysis (PCA). The spectral features were extracted by pseudoinverse linear discriminant analysis (PLDA), improved pseudoinverse linear discriminant analysis (IPLDA), and FIPLDA. Five chrysanthemum tea samples were classified by k-nearest neighbor (KNN) and support vector machines (SVM), respectively. The results show that KNN performed better than SVM in classification, and the classification accuracy of FIPLDA-KNN reached 98.33%. Hence, the combination of NIR spectroscopy with FIPLDA suitable for distinguishing chrysanthemum tea.

Multivariate Curve Resolution Methodology Applied to the ATR-FTIR Data for Adulteration Assessment of Virgin Coconut Oil.

Virgin coconut oil (VCO) is a functional food with important health benefits. Its economic interest encourages fraudsters to deliberately adulterate VCO with cheap and low-quality vegetable oils for financial gain, causing health and safety problems for consumers. In this context, there is an urgent need for rapid, accurate, and precise analytical techniques to detect VCO adulteration. In this study, the use of Fourier transform infrared (FTIR) spectroscopy combined with multivariate curve resolution-alternating least squares (MCR-ALS) methodology was evaluated to verify the purity or adulteration of VCO with reference to low-cost commercial oils such as sunflower (SO), maize (MO) and peanut (PO) oils. A two-step analytical procedure was developed, where an initial control chart approach was designed to assess the purity of oil samples using the MCR-ALS score values calculated on a data set of pure and adulterated oils. The pre-treatment of the spectral data by derivatization with the Savitzky-Golay algorithm allowed to obtain the classification limits able to distinguish the pure samples with 100% of correct classifications in the external validation. In the next step, three calibration models were developed using MCR-ALS with correlation constraints for analysis of adulterated coconut oil samples in order to assess the blend composition. Different data pre-treatment strategies were tested to best extract the information contained in the sample fingerprints. The best results were achieved by derivative and standard normal variate procedures obtaining RMSEP and RE% values in the ranges of 1.79-2.66 and 6.48-8.35%, respectively. The models were optimized using a genetic algorithm (GA) to select the most important variables and the final models in the external validations gave satisfactory results in quantifying adulterants, with absolute errors and RMSEP of less than 4.6% and 1.470, respectively.

Detection of volatile fatty acids in anaerobic digestion system by near infrared spectroscopy

The development of renewable energy technologies, such as anaerobic digestion (AD) systems, is critical to the future of climate-adaptive energy systems. Volatile fatty acid (VFA) is one of the most important intermediates during the AD process, with two effects - inhibiting digestion and promoting CH4 production. Existing techniques for measuring VFA levels are restricted by difficulties in real-time detection, tedious experimentation, laborious processes, and high cost. To achieve the goal of low-cost and high-efficiency, we report the feasibility of evaluating acetic acid, propionic acid, and butyric acid in AD systems using NIRS. We found that the best combinations of preprocessing methods and feature selection were Savitzky-Golay smoothing-multivariate scattering correction-genetic algorithm, Savitzky-Golay smoothing-interval Random Frog, and Savitzky-Golay smoothing-genetic algorithm, respectively. For all three models, the correlation coefficients were above 0.90, and the ratio of performance to deviation was above 3.00. Our results show that the near-infrared spectral model, after preprocessing and feature band extraction, has a good quantitative prediction effect on VFA. This provides technical support for the rapid and non-destructive quantitative analysis of VFA content in AD systems, and contributes to the development of renewable energy technologies.

CEDUP: Using incremental learning modeling to explore Spatio-temporal carbon emission distribution and unearthed patterns at the municipal level

Carbon emissions reduction has become a world consensus. Cities have an essential role to play in addressing emission reductions. However, previous studies have estimated China's municipal-level carbon emissions based on provincial-level emission patterns, and such a top-down carbon emission accounting approach has led to biased results. Therefore, this study employed an incremental learning ensemble model and a Savitzky-Golay algorithm tomeasure carbon emission distribution and unearth patterns at the municipal level based on nighttime light (NTL) and regional development characteristics (GDP, population, patents, industry structure). The performance of the proposed method is substantially better than its counterparts in terms of municipal-level estimation (R-square boosted by 20.64%). This research shows significant difference in carbon emission mechanisms between provinces and cities and demonstrates that carbon emissions are time-continuous. It also shows that per-capita carbon emissions are peaking in many China's cities, except in some heavy industrial cities. Our approach provides accurate and dynamic monitoring of municipal emissions in China.

Application of Spectroscopy to Analyze the Degree of Substitution of Carboxymethyl Cellulose

In this study, the degree of substitution of CMC (carboxymethyl cellulose) was calculated through a titration method, and the data were compared with the measured value of the degree of substitution measured by EDS (energy dispersive X-ray spectroscopy) and the degree of substitution measured by IR (Infrared spectroscopy), and the correlation was analyzed to calculate the degree of substitution of CMC more conveniently. In addition, by identifying the chemical properties of CMC according to the degree of substitution, we tried to provide basic data for the application of CMC in the future. When the degree of substitution of CMC analyzed through EDS analysis was compared with the result of the degree of substitution by the titration method, it was confirmed to have relatively high accuracy, but expensive equipment and remaining NaCl may affect the result. So, it was judged that commercialization may be somewhat difficult. It was easier to analyze the degree of substitution through IR analysis, and more accurate results could be expected than analysis through EDS. Using the Savitzky-Golay algorithm and PCA (principal component analysis), it was possible to identify CMC’s chemical properties and analyze classification according to the degree of substitution. Therefore, it is believed that the application of IR when analyzing the degree of substitution of CMC will simplify the investigation of the relationship between the structure, physical properties, and applications of CMC in the future.

An enhanced principal component analysis method with Savitzky–Golay filter and clustering algorithm for sensor fault detection and diagnosis

Sensors are critical components of heating, ventilation, and air-conditioning systems. Sensor faults can impact control regulations, resulting in an uncomfortable indoor environment and energy wastage. To detect and identify sensor faults quickly, this study proposes an enhanced principal component analysis (PCA) method using the Savitzky–Golay (SG) filter and density-based spatial clustering of applications with noise (DBSCAN) algorithm. First, the DBSCAN algorithm is used to automatically divide the dataset into sub-datasets with different working conditions to reduce the interference information and concentrate the information of each training set. Then, each sub-dataset is smoothed using the SG algorithm to reduce the effects of data fluctuations. The processed dataset is used to build a sub-PCA model that ultimately identifies and locates faults. The proposed strategy is validated using field operating data for 20 air-handling unit (AHU) systems, as obtained from a large commercial building. The fault detection performances of multiple strategies are compared and analysed under different degrees of bias in single AHU and multiple AHU systems. The verification results show that the proposed DBSCAN-SG-PCA model offers significant improvements in fault detection accuracy and fault identification sensitivity over the conventional PCA method. Compared with the SG-PCA model, the proposed model reduces the amount of data required for fault detection by an average of 13.7%, and the Youden index is increased by an average of 0.21. Furthermore, the fault detection accuracy of the proposed model is ±0.7 °C.

Adaptive Savitzky–Golay Filters for Analysis of Copy Number Variation Peaks from Whole-Exome Sequencing Data

Copy number variation (CNV) is a form of structural variation in the human genome that provides medical insight into complex human diseases; while whole-genome sequencing is becoming more affordable, whole-exome sequencing (WES) remains an important tool in clinical diagnostics. Because of its discontinuous nature and unique characteristics of sparse target-enrichment-based WES data, the analysis and detection of CNV peaks remain difficult tasks. The Savitzky–Golay (SG) smoothing is well known as a fast and efficient smoothing method. However, no study has documented the use of this technique for CNV peak detection. It is well known that the effectiveness of the classical SG filter depends on the proper selection of the window length and polynomial degree, which should correspond with the scale of the peak because, in the case of peaks with a high rate of change, the effectiveness of the filter could be restricted. Based on the Savitzky–Golay algorithm, this paper introduces a novel adaptive method to smooth irregular peak distributions. The proposed method ensures high-precision noise reduction by dynamically modifying the results of the prior smoothing to automatically adjust parameters. Our method offers an additional feature extraction technique based on density and Euclidean distance. In comparison to classical Savitzky–Golay filtering and other peer filtering methods, the performance evaluation demonstrates that adaptive Savitzky–Golay filtering performs better. According to experimental results, our method effectively detects CNV peaks across all genomic segments for both short and long tags, with minimal peak height fidelity values (i.e., low estimation bias). As a result, we clearly demonstrate how well the adaptive Savitzky–Golay filtering method works and how its use in the detection of CNV peaks can complement the existing techniques used in CNV peak analysis.

Detection of tomato water stress based on terahertz spectroscopy.

China's tomato cultivation area is nearly 15 thousand km2, and its annual tomato output is about 55 million tons, accounting for 7% of its total vegetable production. Because of the high drought sensitivity of tomatoes, water stress inhibits their nutrient uptake, leading to a decrease in tomato quality and yield. Therefore, the rapid, accurate and non-destructive detection of water status is important for scientifically and effectively managing tomato water and fertilizer, improving the efficiency of water resource utilization, and safeguarding tomato yield and quality. Because of the extreme sensitivity of terahertz spectroscopy to water, we proposed a tomato leaf moisture detection method based on terahertz spectroscopy and made a preliminary exploration of the relationship between tomato water stress and terahertz spectral data. Tomato plants were grown at four levels of water stress. Fresh tomato leaves were sampled at fruit set, moisture content was calculated, and spectral data were collected through a terahertz time-domain spectroscope. The raw spectral data were smoothed using the Savitzky-Golay algorithm to reduce interference and noise. Then the data were divided by the Kennard-Stone algorithm and the sample set was partitioned based on the joint X-Y distance (SPXY) algorithm into a calibration set and a prediction set at a ratio of 3:1. SPXY was found to be the better approach for sample division. On this basis, the stability competitive adaptive re-weighted sampling algorithm was used to extract the feature frequency bands of moisture content, and a multiple linear regression model of leaf moisture content was established under the single dimensions of power, absorbance and transmittance. The absorbance model was the best, with a prediction set correlation coefficient of 0.9145 and a root mean square error of 0.1199. To further improve the modeling accuracy, we used a support vector machine (SVM) to establish a tomato moisture fusion prediction model based on the fusion of three-dimensional terahertz feature frequency bands. As water stress intensified, the power and absorbance spectral values both declined, and both were significantly and negatively correlated with leaf moisture content. The transmittance spectral value increased gradually with the intensification of water stress, showing a significant positive correlation. The SVM-based three-dimensional fusion prediction model showed a prediction set correlation coefficient of 0.9792 and a root mean square error of 0.0531, indicating that it outperformed the three single-dimensional models. Hence, terahertz spectroscopy can be applied to the detection of tomato leaf moisture content and provides a reference for tomato moisture detection.

Nondestructive testing and visualization of compound heavy metals in lettuce leaves using fluorescence hyperspectral imaging

This study evaluated the feasibility of nondestructive testing and visualization of compound heavy metals (cadmium and lead) in lettuce leaves using fluorescence hyperspectral imaging. In addition, a method involving wavelet transform and stepwise regression (WT-SR) was proposed to perform dimensionality reduction of fluorescence spectral data. Fluorescent hyperspectral image acquisition and mathematical analysis were carried out on lettuce leaf samples processed with different compound heavy metal concentrations. The entire lettuce leaf sample was selected as a region of interest (ROI). Savitzky-Golay (SG) algorithm, multivariate scatter correction (MSC), standard normalized variable (SNV), first derivative (1st Der) and second derivative (2nd Der) were used to preprocess the ROI fluorescence spectra. Further, the successive projections algorithm (SPA), the competitive adaptive reweighted sampling (CARS), the iteratively retaining informative variables (IRIV) and variable iterative space shrinkage approach (VISSA), and the wavelet transform combined with stepwise regression (WT-SR) were used to reduce the dimension of spectral data. Finally, the multiple linear regression (MLR) algorithm was used to build the compound heavy metal content detection models. The results showed that the MLR models based on the feature data obtained by 1st Der-WT-SR achieved reasonable performance with Rp2 of 0.7905, RMSEP of 0.0269 mg/kg and RPD of 2.477 for Cd content under wavelet fifth layer decomposition, and with Rp2 of 0.8965, RMSEP of 0.0096 mg/kg and RPD of 3.211 for Pb content under wavelet first layer decomposition. The distribution maps of cadmium and lead contents in lettuce leaves were established using the optimal prediction models. The results further confirmed the great potential of fluorescence hyperspectral technology combined with optimization algorithm for the detection of compound heavy metals.

Microwave Frequency Measurement System Using Fixed Low Frequency Detection Based on Photonic Assisted Brillouin Technique

Microwave frequency measurement system is the chief component of the receiver, which is of indispensable significance in electronic warfare and modern radar. In this work, a simple and novel frequency measurement system that detects at fixed low-frequency point is established, which is implemented by introducing a frequency down-conversion, in conjunction with Brillouin Technique. A swept frequency signal is applied for the purpose of carrying the sideband of unknown signal. The frequency-shifted carrier is realized through the all-optical Brillouin frequency shifting process, which is used to downconvert unknown signals. Eventually, the unknown signal can be obtained by recording the frequency of the sweep signal when the beat signal occurs at a preset fixed frequency of 828 MHz. The proposed approach does not require broadband electronic monitoring equipment, which reduces the pressure of subsequent processing and acquisition. An experiment has been performed to vindicate the feasibility of the measurement system, which shows that a measurement ranges from 1 to 9.5 GHz. Meanwhile, the savitzky-golay algorithm is utilized to improve the frequency resolution to 15 MHz. Furthermore, a large bandwidth measurement range and the expanded uncertainty (95.45% coverage) of 3.604 MHz in the system are also discussed and verified.

Identification of foodborne pathogenic bacteria using confocal Raman microspectroscopy and chemometrics.

Rapid and accurate identification of foodborne pathogenic bacteria is of great importance because they are often responsible for the majority of serious foodborne illnesses. The confocal Raman microspectroscopy (CRM) is a fast and easy-to-use method known for its effectiveness in detecting and identifying microorganisms. This study demonstrates that CRM combined with chemometrics can serve as a rapid, reliable, and efficient method for the detection and identification of foodborne pathogenic bacteria without any laborious pre-treatments. Six important foodborne pathogenic bacteria including S. flexneri, L. monocytogenes, V. cholerae, S. aureus, S. typhimurium, and C. botulinum were investigated with CRM. These pathogenic bacteria can be differentiated based on several characteristic peaks and peak intensity ratio. Principal component analysis (PCA) was used for investigating the difference of various samples and reducing the dimensionality of the dataset. Performances of some classical classifiers were compared for bacterial detection and identification including decision tree (DT), artificial neural network (ANN), and Fisher's discriminant analysis (FDA). Correct recognition ratio (CRR), area under the receiver operating characteristic curve (ROC), cumulative gains, and lift charts were used to evaluate the performance of models. The impact of different pretreatment methods on the models was explored, and pretreatment methods include Savitzky-Golay algorithm smoothing (SG), standard normal variate (SNV), multivariate scatter correction (MSC), and Savitzky-Golay algorithm 1st Derivative (SG 1st Der). In the DT, ANN, and FDA model, FDA is more robust for overfitting problem and offers the highest accuracy. Most pretreatment methods raised the performance of the models except SNV. The results revealed that CRM coupled with chemometrics offers a powerful tool for the discrimination of foodborne pathogenic bacteria.

Forensic Identification of 3,4-Methylendioxy-N-methylamphetamine (MDMA), Ketamine, and Benzodiazepine by Handheld Infrared Spectroscopy and Chemometrics

The development of simple and cost-effective analytical methods is indispensable for law enforcement officials in the rapid screening for large quantities of seized drugs. Handheld infrared spectroscopy has been used with Bayes discrimination analysis (BDA), k nearest neighbor (KNN), support vector machine (SVM), decision-making tree (DT), and random forest (RF) chemometrics to characterize 3,4-methylenedioxy-N-methylamphetamine (MDMA), ketamine, and benzodiazepine samples. Spectral data were pretreated by normalization, multivariate scatter correction, Savitzky-Golay smoothing algorithm, and automatic baseline correction. Principal component analysis (PCA) was employed to reduce the dimensionality of spectral data, while BDA, KNN, SVM, DT, and RF, as supervised pattern recognition methods, were implemented and their parameters were optimized to identify the samples. The results showed that SVM was optimal among the models, followed by DT, KNN, BDA, and RF. The optimal model, based upon polynomial kernel function, achieved an accuracy of 100% (training set) and 100% (test set) for all samples and was used to further distinguish the additives in MDMA, ketamine, and benzodiazepine samples. In the MDMA samples, 42 samples contained caffeine, 23 contained glucose, and 21 contained starch. In the ketamine samples, 34 contained caffeine, 13 contained glucose, and 12 contained phenacetin. In the benzodiazepine samples, estazolam, diazepam, lorazepam, clonazepam, and nitrazepam were identified. 5 samples were misidentified into other categories. This study demonstrated that handheld infrared spectroscopy with chemometrics is a reliable and desirable approach for the rapid and nondestructive identification of MDMA, ketamine, and benzodiazepine with high potential in practical applications.

Rapid diagnosis of diabetes based on ResNet and Raman spectroscopy

Diabetes mellitus is a global public health problem, and the epidemic situation in China is particularly serious. The prevalence of the disease has been increasing in recent years, and the number of patients is the highest in the world. Diabetes has become another chronic non-communicable disease that seriously endangers the health of our people after cardiovascular and cerebrovascular diseases and tumors. In this study, urine sample data were collected from 37 diabetic patients and 37 healthy volunteers using Raman spectroscopy. The collected data were preprocessed using an adaptive iterative reweighted penalized least squares (airPLS) algorithm and a polynomial Savitzky-Golay smoothing algorithm. After extracting features using principal component analysis (PCA) dimensionality reduction algorithm, ResNet, support vector machine (SVM) and linear discriminant analysis (LDA) classification models were selected to classify and identify diabetic patients and healthy controls. The results show that ResNet has the best discrimination effect, and the average accuracy, recall and F1-score can reach 84.28%, 86.20% and 84.02% respectively after five cross-validations, and the area under the subject working characteristic (ROC) curve is 0.93. The experimental results show that the model established in this paper is simple to operate, highly accurate and has good reference value for rapid screening of diabetes.