Ppb-Level Ammonia Sensor for Exhaled Breath Diagnosis Based on UV-DOAS Combined with Spectral Reconstruction Fitting Neural Network.

Ammonia (NH3) in exhaled breath (EB) has been a biomarker for kidney function, and accurate measurement of NH3 is essential for early screening of kidney disease. In this work, we report an optical sensor that combines ultraviolet differential optical absorption spectroscopy (UV-DOAS) and spectral reconstruction fitting neural network (SRFNN) for detecting NH3 in EB. UV-DOAS is introduced to eliminate interference from slow change absorption in the EB spectrum while spectral reconstruction fitting is proposed for the first time to map the original spectra onto the sine function spectra by the principle of least absolute deviations. The sine function spectra are then fitted by the least-squares method to eliminate noise signals and the interference of exhaled nitric oxide. Finally, the neural network is built to enable the detection of NH3 in EB at parts per billion (ppb) level. The laboratory results show that the detection range is 9.50-12425.82 ppb, the mean absolute percentage error (MAPE) is 0.83%, and the detection accuracy is 0.42%. Experimental results prove that the sensor can detect breath NH3 and identify EB in simulated patients and healthy people. Our sensor will serve as a new and effective system for detecting breath NH3 with high accuracy and stability in the medical field.

Sensitive Detection of OCS Using Thermal Conversion Combined with Spectral Reconstruction Filtering Differential Optical Absorption Spectroscopy.

Carbonyl sulfide (OCS) is a toxic gas produced during industrial processes that poses risks to both human health and industrial equipment. Therefore, detecting OCS concentrations plays a crucial role in early hazard warning. This paper presents an online system for detecting OCS at the ppb level using thermal conversion and spectral reconstruction filtering differential optical absorption spectroscopy (SRF-DOAS). First, OCS, which is not suitable for DOAS due to its weak absorption characteristics, is completely transformed into SO2 with strong absorption characteristics under high-temperature conditions. Then, the spectral reconstruction filtering method (SRF) is proposed to eliminate the noise and interference. The core idea of the method is to arrange the spectrum according to the spectral intensity from small to large rather than wavelength, reconstructing the spectrum into a new spectrum with linear characteristics. The reconstructed spectrum can remove noise and interference by linear fitting and retain the characteristic of SO2 oscillation absorption. Next, we demonstrate the ability of the reconstructed spectral method to remove noise and interference by comparing the spectra of the inverse-reconstructed gas mixture and SO2. The relative deviation of 0.88% at 100 ppb and detection limit of 7.26 ppb*m for OCS were obtained using the SRF-DOAS method. Finally, the reliability of the system was confirmed by measurements of OCS concentrations in mixture gas of OCS and air, as well as in human exhaled breath.

Mesenchymal stem cells reverse thymus aging by reprogramming the DNA methylation of thymic epithelial cells.

Background: A decrease in the number and activity of thymic epithelial cells (TECs) is an important factor in thymic degeneration. Mesenchymal stem cells (MSCs) treating thymic ageing is a promising strategy, but the DNA methylation modification mechanism in TECs remains unclear. Methods: Aged rhesus monkeys were treated with MSCs to establish a thymic senescence model, and hematoxylin-eosin (HE) staining, immunofluorescence staining, and ELISA were performed to observe the structure and function of the thymus. TEC aging model and MSCs co-culture system were established to detect DNA methylation modification and transcriptomic changes, correlation analysis between transcription factor methylation and mRNA expression, and q-PCR, immunofluorescence staining, and Western blot were used to identified key genes. Results: MSCs improved the structure and function of thymus in elderly macaque monkeys; reduced the expression levels of ß-Gal, P16, and P21; and increased the activity of aging TECs. There were 501 genes with increased methylation in the promoter region in the treated group compared with the untreated group, among which 23 genes were involved in the negative regulation of cell growth, proliferation and apoptosis, while 591 genes had decreased methylation, among which 37 genes were associated with promoting cell growth and proliferation and inhibiting apoptosis. Furthermore, 66 genes showed a negative correlation between promoter methylation levels and gene transcription; specifically, PDE5A, DUOX2, LAMP1 and SVIL were downregulated with increased methylation, inhibiting growth and development, while POLR3G, PGF, CHTF18, KRT17, FOXJ1, NGF, DYRK3, LRP8, CDT1, PRELID1, F2R, KNTC1 and TRIM3 were upregulated with decreased methylation, promoting cell growth. Conclusion: MSCs improve the structure and function of aged thymus, which involves the regulation of DNA methylation profiles and a decrease in the methylation level of the transcription factor NGF to specifically upregulate KRT17 and FOXJ1 to promote the proliferation of TECs.

Ultrasensitive Online NO Sensor Based on a Distributed Parallel Self-Regulating Neural Network and Ultraviolet Differential Optical Absorption Spectroscopy for Exhaled Breath Diagnosis.

The concentration of fractional exhaled nitric oxide (FeNO) is closely related to human respiratory inflammation, and the detection of its concentration plays a key role in aiding diagnosing inflammatory airway diseases. In this paper, we report a gas sensor system based on a distributed parallel self-regulating neural network (DPSRNN) model combined with ultraviolet differential optical absorption spectroscopy for detecting ppb-level FeNO concentrations. The noise signals in the spectrum are eliminated by discrete wavelet transform. The DPSRNN model is then built based on the separated multipeak characteristic absorption structure of the UV absorption spectrum of NO. Furthermore, a distributed parallel network structure is built based on each absorption feature region, which is given self-regulating weights and finally trained by a unified model structure. The final self-regulating weights obtained by the model indicate that each absorption feature region contributes a different weight to the concentration prediction. Compared with the regular convolutional neural network model structure, the proposed model has better performance by considering the effect of separated characteristic absorptions in the spectrum on the concentration and breaking the habit of bringing the spectrum as a whole into the model training in previous related studies. Lab-based results show that the sensor system can stably achieve high-precision detection of NO (2.59-750.66 ppb) with a mean absolute error of 0.17 ppb and a measurement accuracy of 0.84%, which is the best result to date. More interestingly, the proposed sensor system is capable of achieving high-precision online detection of FeNO, as confirmed by the exhaled breath analysis.

Quantitative risk analysis of sediment heavy metals using the positive matrix factorization-based ecological risk index method: a case of the Kuye River, China.

Identifying the sources of heavy metals (HMs) in river sediments is crucial to effectively mitigate sediment HM pollution and control its associated ecological risks in coal-mining areas. In this study, ecological risks resulting from different pollution sources were evaluated using an integrated method combining the positive matrix factorization (PMF) and the potential ecological risk index (RI) model. A total of 59 sediment samples were collected from the Kuye River and analyzed for eight HMs (Zn, Cr, Ni, Cu, Pb, As, Cd, and Hg). The obtained results showed that the sediment HM contents were higher than the corresponding soil background values in Shaanxi Province. The average sediment Hg content was 3.42 times higher than the corresponding background value. The PMF results indicated that HMs in the sediments were mainly derived from industrial, traffic, agricultural, and coal-mining sources. The RI values ranged from 26.15 to 483.70. Hg was the major contributor (75%) to the ecological risk in the vicinity of the Yanjiata Industrial Park. According to the PMF-based RI model, coal-mining activities exhibited the strongest impact on the river ecosystem (48.79%), followed, respectively, by traffic (34.41%), industrial (12.70%), and agricultural (4.10%) activities. These results indicated that the major anthropogenic sources contributing to the HM contents in the sediments are not necessarily those posing the greatest ecological risks. The proposed integrated approach in this study was useful in evaluating the ecological risks associated with different anthropogenic sources in the Kuye River, providing valuable suggestions for reducing sediment HM pollution and effectively protecting river ecosystems.

Mesenchymal Stem Cells: A Promising Treatment for Thymic Involution.

The thymus is the main immune organ in the body. However, the thymus gradually degenerates in early life, leading to a reduction in T-cell production and a decrease in immune function. Mesenchymal stem cells (MSCs) are a promising alternative for the treatment of thymus senescence due to their homing ability to the site of inflammation and their paracrine, anti-inflammatory, and antioxidant properties. However, the heterogeneity, difficulty of survival in vivo, short residence time, and low homing efficiency of the injected MSCs affect the clinical therapeutic effect. This article reviews strategies to improve the efficacy of mesenchymal stem cell therapy, including the selection of appropriate cell doses, transplantation frequency, and interval cycles. The survival rate of MSCs can be improved to some extent by improving the infusion mode of MSCs, such as simulating the in vivo environment, applying the biological technology of hydrogels and microgels, and iron oxide labeling technology, which can improve the curative effect and homing of MSCs, promote the regeneration of thymic epithelial cells, and restore the function of the thymus.

[Occurrence Characteristics, Sources, and Toxicity Risk Analysis of Polycyclic Aromatic Hydrocarbons in Typical Rivers of Northern Shaanxi Mining Area, China].

The aim of this study was to assess the occurrence level, spatial distribution, pollution source, and ecological risk of polycyclic aromatic hydrocarbons (PAHs) in the Kuye River of the northern Shaanxi mining area. In total, 16 priority PAHs were quantitatively detected at 59 sampling sites using a high-performance liquid chromatography-diode array detector in series with a fluorescence detector. The results showed that the ρ(ΣPAHs) in the Kuye River ranged from 50.06 to 278.16 ng·L-1, with an average value of 128.22 ng·L-1. The PAHs monomer concentrations ranged from 0 to 121.22 ng·L-1, of which Chrysene had the highest concentration, with average values of 36.58 ng·L-1, respectively, followed by benzo(a)anthracene and phenanthrene. The detection rate of each monomer was more than 70%, of which 12 monomers revealed detection rates of 100%. In addition, the 4-ring PAHs showed the highest relative abundance in the 59 samples, ranging from 38.59% to 70.85%. The PAHs concentrations revealed significant spatial variation in the Kuye River. Moreover, the highest PAHs concentrations were mainly observed in coal mining, industrial, and densely populated areas. Compared with those in other rivers in China and worldwide, the PAHs concentrations in the Kuye River showed a medium pollution level. On the other hand, the positive definite matrix factorization (PMF) and diagnostic ratios were used to quantitatively assess the source apportionment of PAHs in the Kuye River. The results showed that coking and petroleum emissions, coal combustion, fuel-wood combustion, and automobile exhaust emissions contributed to the PAHs concentrations in the industrial areas of the upper reach by 34.67%, 30.62%, 18.11%, and 16.60%, and coal combustion, fuel-wood combustion, and automobile exhaust emissions contributed in the downstream residential areas by 64.93%, 26.20%, and 8.86%. In addition, the results of the ecological risk assessment showed low ecological risks of naphthalene and high ecological risks of benzo(a)anthracene, respectively, whereas the remaining monomers revealed medium ecological risk. Among the 59 sampling sites, only 12 belonged to low ecological risk areas, whereas the remaining sampling sites were at medium to high ecological risks. Moreover, the water area near the Ningtiaota Industrial Park showed a risk value close to the high ecological risk threshold. Therefore, it is urgent to formulate prevention and control measures in the study region.

Manipulation of the photoluminescence of lead halide perovskite quantum dots with mechanically reconfigurable 3D photonic crystals.

Reconfigurable 3D photonic crystals (3DPCs) are promising for dynamic emission devices, owing to their unique properties. Here, we integrated the perovskite quantum dot film together with 3D reconfigurable photonic crystals (PCs) to form quantum dot/photonic crystal heterostructures and investigated their interactions at their interfaces. The photonic bandgaps of the presented 3DPCs can be dynamically tuned by heating and applying external mechanical forces, and they can be stably fixed in the intermediate states. By tuning the photonic bandgaps of the 3DPCs, a maximum photoluminescence (PL) enhancement of 11 times that of CsPb(I/Br)3 quantum dots has been achieved. It has been revealed that the combined effects of increased density of photon states and the greatly confined and enhanced electric field on the upper surface of 3DPCs contribute to the enhanced Purcell effect, which in turn leads to the enhanced photoluminescence.

An adaptive absorption spectroscopy with adjustable moving window width for suppressing nonlinear effects in absorbance measurements.

Absorption spectroscopy based on Lambert-Beer law has been widely used in material structure analysis, research in chemical reaction kinetics, and exploration of various physicochemical reaction mechanisms. However, serious nonlinearity between absorbance and measured concentration can occur in actual measurements. The idea of moving window is first introduced into the field of spectral nonlinearity in the paper. Combining with the characteristic absorption spectra of the substances to be measured, we propose an adaptive absorption spectroscopy (A-AS) with adjustable moving window parameters to effectively suppress the nonlinear effects in absorbance measurements. The validity of this method is verified by taking the differential optical absorption spectroscopy to detect SO2 as an example. The 210-230 nm characteristic absorption band is traversed and divided by the moving window with adjustable parameters, and the estimated coefficient (k-value) of each band is calculated. On this basis, all k-values are initially and secondly screened to obtain the optimal kbest, and then the optimal concentration value is obtained by inversion. Compared with the broad-band method and narrow-band method, it shows excellent performance that the maximum error and standard deviation of A-AS is only 1.3% and 3.8 in the entire concentration range, suggesting good linearity and stability in both high and low concentration environments. Therefore, it is inferred that A-AS is universally adaptable and enables dynamic linear measurements over wide concentration range.

Distribution, sources, and ecological risk assessment of polycyclic aromatic hydrocarbons in surface water in the coal mining area of northern Shaanxi, China.

This study investigated the spatial distribution, pollution source, and ecological risk of polycyclic aromatic hydrocarbons (PAHs) in the Kuye River, which is a typical river in the mining area of China, 16 priority PAHs were quantitatively detected at 59 sampling sites by high-performance liquid chromatography-diode array detector-fluorescence detector. The results showed that the ∑PAHs concentrations in the Kuye River were in the range of 50.06-278.16 ng/L. The PAHs monomer concentrations were in the range 0-121.22 ng/L, of which chrysene had the highest average concentration (36.58 ng/L), followed by benzo[a]anthracene and phenanthrene. In addition, the 4-ring PAHs showed the highest relative abundance in the 59 samples, ranging from 38.59 to 70.85%. Moreover, the highest concentrations of PAHs were mainly observed in coal mining, industrial, and densely populated areas. On the other hand, according to the diagnostic ratios and positive matrix factorization (PMF) analysis, it can be concluded that coking/petroleum sources, coal combustion, vehicle emission, and fuel-wood combustion contributed to the PAHs concentrations in the Kuye River by 37.91%, 36.31%, 13.93%, and 11.85%, respectively. In addition, the results of the ecological risk assessment indicated that benzo[a]anthracene had a high ecological risk. Among the 59 sampling sites, only 12 belong to low ecological risk areas, and others were at medium to high ecological risks. The current study provides data support and a theoretical basis to effectively manage pollution sources and ecological environment treatment in mining areas.

Synthetic flavagline derivative 1-chloroacetylrocaglaol promotes apoptosis in K562 erythroleukemia cells through miR-17-92 cluster genes.

Chronic myeloid leukemia accounts for human deaths worldwide and could enhance sevenfold by 2050. Thus, the treatment regimen for this disorder is highly crucial at this time. Flavaglines are a natural class of cyclopentane benzofurans exhibiting various bioactivities like anticancer action. Despite the antiproliferative activity of flavaglines against diverse cancer cells, their roles and mechanism of action in chronic myeloid leukemia (CML) remain poorly understood. Thus, this study examines the antiproliferative effect of a newly synthesized flavagline derivative, 1-chloracetylrocaglaol (A2074), on erythroleukemia K562 cells and the zebrafish xenograft model. The study revealed that A2074 could inhibit proliferation, promote apoptosis, and boost megakaryocyte differentiation of K562 cells. This flavagline downregulated c-MYC and miR-17-92 cluster genes, targeting upregulation of the apoptotic protein Bcl-2-like protein 11 (BIM). The work uncovered a critical role of the c-MYC-miR-17-92-BIM axis in the growth and survival of CML cells.

The Flavagline Compound 1-(2-(dimethylamino)acetyl)-Rocaglaol Induces Apoptosis in K562 Cells by Regulating the PI3K/Akt/mTOR, JAK2/STAT3, and MAPK Pathways.

Purpose: Chronic myelogenous leukemia (CML) is a hematological malignancy with increased proliferation of cells of the myeloid series. This can disrupt normal hematopoiesis. The 1-(2-(dimethylamino)acetyl)-rocaglaol (MQ-16) is a new synthetic flavagline compound that showed promising activity in chronic myeloid leukemia K562 cells. This study aims to analyze the underlying mechanisms of MQ-16 against CML. Methods: Growth, cell cycle progression, and apoptosis were assessed in K562 cells following MQ-16 exposure by MTT assay and flow cytometry. The effect of MQ-16 on DNA strands between nucleosomes was examined by 1% agarose gel electrophoresis. PI3K/Akt/mTOR, JAK2/STAT3, and mitogen-activated protein kinase (MAPK) pathway-related proteins were detected in MQ-16-treated K562 cells by Western blot. Results: MQ-16 significantly inhibited the proliferation of K562 cells and arrested the cell cycle at the G2/M phase in a time- and concentration-dependent manner. MQ-16 induced mitochondria-dependent apoptosis by downregulating the anti-apoptotic proteins Bcl-2 and Bcl-xL and induced time- and concentration-dependent DNA fragmentation. In addition, MQ-16 affected the expression of PI3K/Akt/mTOR, JAK2/STAT3, and MAPK pathway-related proteins. Conclusion: In summary, MQ-16 appears to be a promising chemotherapeutic drug for treating CML.

Three-dimensional chiral metasurfaces for circular-polarized anomalous beam steering.

Metasurfaces offer unprecedented possibilities for developing versatile ultracompact photonic devices with unique functions, e.g., for linear- or circular-polarized light beam steering. Here we report a three-dimensional (3D) chiral metasurface for phase controlling and beam steering, which consists of periodically arranged double-layer circular arc chiral nanostructures. By tuning the central angle of the lower circular arc, the left- and right-circularly polarized light (LCP and RCP) induce different spatial phases, which have been designed as a beam steering device to realize the abnormal reflection of LCP and the mirror reflection of RCP in the near-infrared (NIR) spectrum from 900 nm to 1150 nm, providing a potential device for chiral molecule detection.

Identification and quantification of adulterated honey by Raman spectroscopy combined with convolutional neural network and chemometrics.

In this study, Raman spectroscopy combined with convolutional neural network (CNN) and chemometrics was used to achieve the identification and quantification of honey samples adulterated with high fructose corn syrup, rice syrup, maltose syrup and blended syrup, respectively. The shallow CNNs utilized to analyze honey mixed with single-variety syrup classified samples into four categories by the adulteration concentration with more than 97% accuracy, and the general CNN model for simultaneously detecting honey adulterated with any type of syrup obtained an accuracy of 94.79%. The established CNNs had the best performance compared with several chemometric classification algorithms. In addition, partial least square regression (PLS) successfully predicted the purity of honey mixed with single syrup, while coefficients of determination and root mean square errors of prediction were greater than 0.98 and less than 3.50, respectively. Therefore, the proposed methods based on Raman spectra have important practical significance for food safety and quality control of honey products.

Photoluminescence enhancement in monolayer MoS2 and self-assembled 3D photonic crystal heterostructures.

Self-assembled photonic crystals (PCs) have promising applications in enhancing and directional manipulation of the photoemission due to their photonic bandgaps. Here, we employed self-assembled 3D polystyrene PCs to enhance the photoluminescence (PL) of monolayer molybdenum disulfide (MoS2). Through tuning the photonic bandgap of the polystyrene crystals to overlap with the direct emission band of monolayer MoS2, the MoS2/3D-PC heterostructure showed a maximum 12-fold PL enhancement, and Rabi splitting was also observed in the reflection spectrum. The heterostructure is expected to be useful in nanophotonic emitting devices.

Quantitative analysis of blended corn-olive oil based on Raman spectroscopy and one-dimensional convolutional neural network.

Blended vegetable oil is a vital product in the vegetable oil market, and quantifying high-value vegetable oil is of great significance to protect the rights and interests of consumers. In this study, we established a one-dimensional convolutional neural network (1D CNN) quantitative identification model based on Raman spectra to identify the amount of olive oil in a corn-olive oil blend. The results show that the 1D CNN model based on 315 extended average Raman spectra can quantitatively identify the content of olive oil, with R2p and RMSEP values of 0.9908 and 0.7183 respectively. Compared with partial least squares regression (PLSR) and support vector regression (SVR), although the index is not optimal, it provides a new analytical method for the quantitative identification of vegetable oil.

Circular RNA circ_0040823 inhibits the proliferation of acute myeloid leukemia cells and induces apoptosis by regulating miR-516b/PTEN.

OBJECTIVE: Endogenous circular RNAs (circRNAs) and microRNAs (miRNAs) have been shown to regulate the pathogenesis of acute myeloid leukemia (AML). The current study aimed to identify the role of circRNA 0040823 (circ_0040823) in AML. METHODS: Microarray datasets were analyzed to identify differentially expressed circRNAs in AML patients. Peripheral blood samples were obtained from healthy volunteers and AML patients for the measurement of circ_0040823 and miR-516b levels. The overexpression or knockdown of a target gene in AML cells was achieved by the transfection with lentiviral vectors or small interfering RNAs. BALB/c nude mice were inoculated with AML cells and monitored for tumor growth. Dual-luciferase reporter assay, RNA immunoprecipitation, and RNA pull-down assay were used to determine the binding relationship between circRNA and miRNA. RESULTS: circ_0040823 was significantly downregulated in AML patients and leukemia cells. Overexpression of circ_0040823 inhibited AML cell proliferation, and induced apoptosis and cell cycle arrest. Upregulation of circ_0040823 also repressed the growth of xenograft tumors in vivo. circ_0040823 acted as a miR-516b sponge and regulated key cellular events in leukemia cells via downregulating miR-516b. Moreover, tumor suppressor phosphatase and tensin homolog (PTEN) was a downstream target of miR-516b. The inhibition of miR-516b impaired the proliferation capacity of leukemia cells and induced apoptosis, while PTEN deficiency attenuated these effects. CONCLUSION: This study showed that circ_0040823 inhibited proliferation and induced apoptosis of AML cells by sponging miR-516b, thereby diminishing the regulatory effect of miR-516b on PTEN. These findings identified circ_0040823/miR-516b/PTEN as a new therapeutic target for AML.

Analysis of the mechanism for enhanced pyrene biodegradation based on the interactions between iron-ions and Rhodococcus ruber strain L9.

A slow degradation rate and low transformation efficiency are the main problems in the biodegradation of polycyclic aromatic hydrocarbons (PAHs). This study selected pyrene as the target PAH to investigate the effect of ferrous ion and ferric ion on pyrene degradation. The driving effect and mechanism, including the interaction between pyrene and iron ions and the bacterial physiological response during the biodegradation process by Rhodococcus ruber strain L9, were investigated. The results showed that iron ions did not enhance bacterial growth but improved bacteria's pyrene removal capacity, contributing to the total efficiency of pyrene biodegradation. The process started with an initial formation of "cation-π" between Fe (III) and pyrene, which subsequently drove the pyrene removal process and accelerated the bacterial metabolic process. Moreover, a significant increase in the protein concentration, catechol dioxygenase (C12O and C23O) activities, and intracellular protein regulation in crude enzyme solution indicate a positive response of the bacteria during the iron ion-enhanced pyrene degradation process.

Mesenchymal stem cell carriers enhance antitumor efficacy induced by oncolytic reovirus in acute myeloid leukemia.

Chemotherapy is the main treatment for acute myeloid leukemia (AML), but the therapeutic efficacy is modest, and most commonly manifests as relapse from remission. Thus, improving long-term AML survival is a crucial clinical challenge. In recent years, oncolytic virotherapy has provided an alternative approach for AML treatment. The use of oncolytic reoviruses has been explored in more than 30 clinical trials for safety and feasibility issues. However, like other oncolytic viruses, neutralizing antibodies (NAbs) reduce therapeutic efficacy. To tackle this problem, human umbilical cord mesenchymal stem cells (hUC-MSCs) were used to deliver reovirus using in vitro and in vivo models. Human UC-MSCs were successfully loaded with reovirus, without impairing biological function.We also observed in vitro protective effects of hUC-MSCs on reovirus in the presence of NAbs. In the immunocompromised AML mouse model, hUC-MSCs effectively carried reoviruses to tumor lesions and significantly prolonged the survival of AML xenografts in mice in the presence of a high titer anti-reovirus antibody (p = 0.001). However, reovirus-induced activation of AKT, stress-activated protein kinase/c-Jun N-terminal kinase (SAPK/JNK), and NF-κB signaling led to the maintenance of intrinsic migratory properties and secretion of pro-inflammatory cytokines from hUC-MSCs, particularly CXCL10. In immuno-competent AML mice, MSCs carrying reovirus triggered immune responses, and eventually inhibited tumor growth. Therefore, these results suggest that MSCs as carriers of oncolytic reoviruses can enhance the antitumor efficacy of virotherapy.

Exploration of total synchronous fluorescence spectroscopy combined with pre-trained convolutional neural network in the identification and quantification of vegetable oil.

In order to distinguish different vegetable oils, adulterated vegetable oils, and to identify and quantify counterfeit vegetable oils, a method based on a small sample size of total synchronous fluorescence (TSyF) spectra combined with convolutional neural network (CNN) was proposed. Four typical vegetable oils were classified by three ways of fine-tuning the pre-trained CNN, the pre-trained CNN as a feature extractor, and traditional chemometrics. The pre-trained CNN was combined with support vector machines to distinguish adulterated sesame oil and counterfeit sesame oil separately with 100% correct classification rates. The pre-trained CNN combined with partial least square regression was used to predict the level of counterfeit sesame oil. The coefficient of determination for calibration (Rc2) values were all greater than 0.99, and the root mean square errors of validation were 0.81% and 1.72%, respectively. These results show that it is feasible to combine TSyF spectra with CNN for vegetable oil identification.