Number Of Samples Research Articles

Simultaneous Genotyping of Three SNVs, rs5471, rs5472, and rs2000999 Involved in Serum Haptoglobin Levels by Fluorescent Probe-Based Melting Curve Analysis.

Haptoglobin (Hp) is a hemoglobin-binding acute-phase serum protein. Several single nucleotide variations (SNVs) within the Hp gene (HP) or Hp-related protein gene (HPR), such as rs5471 (A>C) and rs5472 (A>G) in HP promoter region and rs2000999 (G>A) in intron 2 of HRP, are suggested to correlate with the serum Hp levels. To determine these three SNVs simultaneously, a genotyping assay based on duplex dual-labeled fluorescent probes was developed. The method was then validated by analyzing genomic DNA from 121 Ghanaian and two Japanese subjects who had been previously genotyped for rs5471, rs5472, and rs2000999. Both rs5471 and rs5472 could be determined as haplotypes with a single FAM-labeled fluorescent probe, and rs2000999 could be genotyped with a HEX-labeled fluorescent probe. The results obtained with the present method were consistent with the previous results except for those of three Ghanaian subjects. All three subjects appear to have multiple HPR copy number variants characteristic of African populations, which may have led to incorrect results during previous genotyping. This method allows us to genotype these three SNVs in a relatively large number of samples, especially in African populations where rs5471 is uniquely distributed.

Enhancing the accuracy of blockage detection in centrifugal pump using the majority voting classifier on an unbalanced dataset

Blockage in a centrifugal pump can adversely impact its performance. In this study, an experimental facility is developed to simulate three types of blockages (suction, discharge, and simultaneous suction and discharge). To classify these faults and detect their severity, a methodology involving the application of majority voting classifier (MVC) to the pump’s discharge pressure signals is presented. An unbalanced dataset is constructed, where the number of samples for a specific blockage condition decreases with increasing fault severity. Statistical features, entropy features, and entropies meta features are extracted from the signal and ranked using XGBoost and minimum redundancy maximum relevance (MRMR). Subsequently, the optimal features are selected based on the best performing model (Linear Discriminant Analysis) among ten different models. Best four models are selected and ensembled to form MVC. Results show that MVC achieves an accuracy of 89.90% and 88.26% for features selected by XGBoost and MRMR, respectively. Finally, the unbalanced dataset is balanced using synthetic minority oversampling and it is shown that MVC achieves an accuracy of 100% on this balanced dataset for the features selected using both approaches.

Ensemble Learning for Stellar Classification and Radius Estimation from Multimodal Data

Abstract Stellar classification and radius estimation are crucial for understanding the structure of the universe and stellar evolution. With the advent of the era of astronomical big data, multimodal data are available and theoretically effective for stellar classification and radius estimation. A problem is how to improve the performance of this task by jointly using the multimodal data. However, existing research primarily focuses on using single-modal data. To this end, this paper proposes a model, Multi-Modal SCNet (MMSCNet), and its ensemble model Multimodal Ensemble for Stellar Classification and Regression (MESCR) for improving stellar classification and radius estimation performance by fusing two modality data. In this problem, a typical phenomenon is that the sample numbers of some type stars are evidently more than others. This imbalance has negative effects on model performance. Therefore, this work utilize a weighted sampling strategy to deal with the imbalance issues in MESCR. Some evaluation experiments are conducted on a test set for MESCR and the classification accuracy is 96.1\%, the radius estimation performance Mean of Absolute Error (MAE) and $\sigma$ are 0.084 dex and 0.149 \(R_{\odot}\) respectively. Moreover, we assessed the uncertainty of model predictions, confirming good consistency within a reasonable deviation range. Finally, we applied our model to 50,871,534 SDSS stars without spectra and published a new catalog.

Malware Detection Based on API Call Sequence Analysis: A Gated Recurrent Unit–Generative Adversarial Network Model Approach

Malware remains a major threat to computer systems, with a vast number of new samples being identified and documented regularly. Windows systems are particularly vulnerable to malicious programs like viruses, worms, and trojans. Dynamic analysis, which involves observing malware behavior during execution in a controlled environment, has emerged as a powerful technique for detection. This approach often focuses on analyzing Application Programming Interface (API) calls, which represent the interactions between the malware and the operating system. Recent advances in deep learning have shown promise in improving malware detection accuracy using API call sequence data. However, the potential of Generative Adversarial Networks (GANs) for this purpose remains largely unexplored. This paper proposes a novel hybrid deep learning model combining Gated Recurrent Units (GRUs) and GANs to enhance malware detection based on API call sequences from Windows portable executable files. We evaluate our GRU–GAN model against other approaches like Bidirectional Long Short-Term Memory (BiLSTM) and Bidirectional Gated Recurrent Unit (BiGRU) on multiple datasets. Results demonstrated the superior performance of our hybrid model, achieving 98.9% accuracy on the most challenging dataset. It outperformed existing models in resource utilization, with faster training and testing times and low memory usage.

Metabolomic changes in preterminal serum samples of rhesus macaques exposed to two different lethal doses of total-body gamma-radiation

Exposure to ionizing radiation induces cellular and molecular damage leading to a cascade of events resulting in tissue and organ injury. Our study strives to characterize and validate metabolomic changes in preterminal stage (immediately prior to death) samples collected from rhesus macaques lethally irradiated with one of two different doses of radiation. Peripheral blood samples were collected pre-exposure, post-exposure, and at the preterminal stage of nonhuman primates (NHPs that did not survive exposure with 7.2 Gy or 7.6 Gy total-body radiation (LD60-80/60)). We analyzed global metabolomic alterations using ultra-performance liquid chromatography (UPLC) quadrupole time-of-flight mass spectrometry (QTOF-MS) in serum samples collected at various timepoints in relation to radiation exposure. The goal of this study was to validate the metabolic shifts present in samples collected just prior to death, which were reported earlier in a preliminary study with a limited number of samples and a single dose of radiation. Here, we demonstrate that radiation exposure induced significant time-dependent metabolic alterations compared with pre-exposure samples. We observed significant metabolite dysregulation in animals exposed to 7.6 Gy compared to 7.2 Gy. Greater metabolic disruption was observed in the preterminal groups than all of the other post-irradiation timepoints in both cohorts. Metabolomic shifts in these preterminal groups also revealed consistent disturbances in sphingolipid metabolism, steroid hormone biosynthesis, and glycerophospholipid metabolism pathways. Overall, the sphingolipid metabolism pathway appears to be representative of the preterminal phenotype, confirming the results of our preliminary study. These results offer important and novel insights for identification and validation of biomarkers for lethality, and such observations would be valuable for triage during a radiological/nuclear mass casualty scenario.

Levels of chlorpyrifos, ethion, lambda-cyhalothrin, pendimethlin and triazophos pesticide residues in vegetables

Keeping in view the adverse effects of pesticides, due to their excessive uses in crops and storages, a representative study was conducted to evaluate the presence of pesticide residues in vegetables for most frequently used pesticides in the state of Haryana in India. In all, a variety of vegetables were collected as samples from farmers as well as markets located in twenty one districts of Haryana, India for the period of eight years. The extraction of pesticides from vegetable samples was done using QuEChERS method and analysis for commonly used pesticides was done using Gas chromatography–mass spectrometry (GCMS). The analysis of samples showed that approximately one-third of samples were found with detectable presence of pesticide residues. More importantly, the presence of some banned or withdrawn pesticides such as DDT isomers, endosulphan. Furthermore, approximately 40 % of the analyzed samples contained pesticide residues at or below maximum residue level (MRLs) and 16 % of analyzed samples contained pesticide residues above MRLs in terms of commodity groups. Furthermore, residues of Chlorpyrifos were found most often exceeding the Food Safety and Standards Authority of India (FSSAI). Ethion, lambda-cyhalothrin, pendimethlin and triazophos were other pesticides with detected value higher than specified MRL values. Pesticides occurrence in large number of samples need to be "alarming and forcing matter to take appropriate action" to the cause of public health. This study may provide new directions for the application, management and control of pesticides used by farmers.

A Small-Sample Classification Strategy for Extracting Fractional Cover of Native Grass Species and Noxious Weeds in the Alpine Grasslands.

The fractional cover of native grass species (NGS) and noxious weeds (NW) provides a more comprehensive understanding of grassland health in the alpine grasslands. However, coverage extraction of NGS and NW from satellite hyperspectral imagery can be challenging due to the small spectral and spatial feature difference, insufficient training samples, and the lack of effective fractional cover extraction methods. In this research, firstly, a feature optimization method is proposed to optimize the difference feature between NGS and NW. Secondly, a spectral-spatial constrained re-clustering training sample extension method (SSCTSE) is proposed to increase the number of training samples. Thirdly, a composite three-kernel SVM method (CTK-SVM) is developed to produce fractional cover maps of NGS and NW. The experimental results show that (1) the feature optimization method is effective in preserving the spectral and spatial difference features while eliminating invalid features; (2) the SSCTSE algorithm is capable of significantly increasing the number of training samples; (3) the fractional cover maps of NGS and NW are produced with the CTK-SVM method with overall accuracies of approximately 65%, and the RMSEs of NGS and NW are approximately 16% and 11%, respectively. The results provide a foundation for the fractional cover extraction of different grass species in alpine grasslands based on satellite hyperspectral imagery.

Determining the Soil Particle Shape by Use of Dynamic Image Analysis

In Soil Science, it is well established that the geometry of individual soil particles, including their shape and size, has a significant impact on a number of key physical properties of the soil and on the processes involved with it. The shape and size of soil particles directly impact aspects such as permeability, water retention, stability of soil aggregates and the availability of nutrients for plants. Therefore, accurate determination of these characteristics of soil particles becomes extremely important for understanding and effective soil management. Nowadays, dynamic image analysis (DIA) is emerging as an exceptionally versatile and effective technique that enables precise determination of particle characteristics, including shape and size, in a dynamic and non-invasive manner. DIA enables the automatic analysis of thousands of microscopic images, allowing rapid and accurate study of the morphology of soil particles at micro and macro scales. The use of dynamic image analysis in soil research provides insight into the complex structures of soil particles and better prediction of their impact on various soil processes. Moreover, DIA enables the examination of a large number of samples in a short time, which contributes to the efficiency and precision of soil testing. Therefore, dynamic image analysis is becoming a widely used technique for determining particles' characteristics, such as shape and size. In the context of an ever-increasing awareness of the sustainable use of natural resources and the need to optimize agricultural and engineering practices, dynamic image analysis is becoming an indispensable tool for soil scientists, environmental scientists and engineers. Its versatile use can contribute to further expanding our knowledge about soil and developing innovative solutions for sustainable soil and environmental management.

Preparation and Application of Polyclonal Antibodies for the Rapid Detection of Actinidia Chlorotic Ringspot-Associated Virus

Actinidia chlorotic ringspot-associated virus (AcCRaV, Emaravirus actinidiae) is prevalent in Chinese kiwifruit, leading to substantial yield reduction. The intricate nature of symptoms presents diagnostic challenges, underscoring the necessity for a rapid and accurate detection method that facilitates effective control. In this investigation, AcCRaV isolates from key kiwi-producing regions in Sichuan province were collected and analyzed, with representative strains chosen as experimental materials. Primers targeting the nucleoprotein gene of AcCRaV were designed, and their codon usage was optimized to enhance performance. Various serological methods utilizing polyclonal antibodies were developed, including ELISA, dot immunobinding assay, and AcCRaV-specific gold immunochromatographic bands (AcCRaV-GICS). Field samples exhibited high specificity and sensitivity when tested using these methods. Furthermore, the results obtained from a large number of field samples are consistent with those derived from RT-PCR analysis, further validating the applicability of our approach. A detection method capable of handling a large volume of field samples infected with AcCRaV is currently lacking; thus, our system construction provides an important reference for addressing this gap.

Lavender essential oil market integrity: A comprehensive study of commercial Lavandula angustifolia essential oils adulteration and assessment of industrial standards

Adulteration of essential oils (EOs) is a common and unethical practice that affects both consumers and honest suppliers and harms market integrity. Despite the awareness of this problem, scarce scientific studies are described in the scientific literature. Therefore, 55 samples of lavender EOs were purchased and extensively analyzed to the trace levels, including chromatographic profile, quantitative analysis, and chiral profile. Of the 51 samples labeled as Lavandula angustifolia, 51 % were genuine, 6 % were cheaper lavandin EO, 14 % were fortified with synthetic additives and over 29 % of the samples were adulterated with both lavandin EO and synthetics. Usually, adulterated EOs were cheaper than genuine products, still some expensive products were also falsified. The most common synthetic adulterants were synthetic linalool and linalyl acetate, 3,5,5-trimethylhexyl acetate, α-terpinyl acetate, and dipropylene glycol. Only four of the studied samples were diluted with a medium not fully visible in the recorded GC profile. Industry standards were assessed to eliminate adulterated EOs and ISO 3515 was found to be impractical and too strict, but Ph. Eur. demonstrated its efficiency in eliminating adulterated samples without rejecting many genuine ones.PCA enabled the differentiation of lavender EOs containing synthetics from those of purely natural origin, and in the latter group, to distinguish L. angustifolia from lavandin. Some samples had an unusual composition suggesting that they were artificial blends of synthetics and some terpene fractions of non-lavender origin.This is the first such extensive study of adulteration of lavender EOs, regarding both the number of commercial samples as well as implications on their biological activity. The adulteration via dilution influenced the antimicrobial activity of the essential oils. In conclusion, adulteration is a significant problem in the market, and a large share of products on the market is somehow counterfeited, not only the cheapest products.

Preparation of Dysprosium(III)-Metal Organic Framework Nanofiber for Exosome Capture and Biomarker Discovery toward Liver Disease.

As an emerging source for liquid biopsy, exosomes hold significant promise for clinical diagnosis. However, commonly used exosome isolation methods (e.g., ultracentrifugation) suffer from low throughput for a large number of clinical samples. Herein, a dysprosium-metal organic framework was synthesized and doped with nanofibers by electrospinning for efficient capture of exosomes from body fluid. With the integration of multichannel of pipet or robot automatic workstation, high throughput exosome isolation can be achieved with clinical samples with high reproducibility. To evaluate the clinical value of the developed method, urinary exosomes were enriched from 34 liver disease samples of different stages for the profiling of metabolites by mass spectrometry. The results showed that HCC, cirrhosis, and healthy controls can be significantly differentiated by the Random Forest classification model. The dysprosium-metal organic framework has promising applications in exosome-based liquid biopsy for large-scale clinical disease diagnosis.

The Influence of Leadership Style and Cognitive Attitudes on Patient Safety Culture With Coordination as a Moderation Variable in Inpatient Nurses at Tzu Chi Hospital

This research is based on the results of initial observations which indicate that several safety incident problems will occur in the year 2022. The aim of this research is to reveal empirically the influence of leadership style and cognitive attitudes on patient safety culture with coordination as a moderator. This type of research is quantitative with a cross sectional study design. The population used is nurses who work in inpatient installations. The sampling technique is non-probability sampling, with perm = calculation of the number of samples using Slovin's formula with an error rate of 10%, so that a total sample of 64 respondents was found. The data collection technique uses a questionnaire, and the analysis method uses three box method analysis and PLS-SEM with the help of the Smart-PLS program. The results of the analysis prove that leadership style, cognitive attitude and coordination simultaneously and partially influence patient safety culture, and cognitive attitude is the most dominant variable capable of improving patient safety culture. Coordination cannot moderate the influence of leadership style on patient safety culture, but coordination can weaken the influence of cognitive attitudes on patient safety culture.

Simplified efficiency calibration methods for scintillation detectors used in nuclear remediation

Our study introduces innovative methods to address waste reduction and enhance resource efficiency in nuclear cleanup processes. We focus on sustainable nuclear technology, aligning with goals for cleaner production and environmental protection. We developed two novel methods, termed “oversimplified” and “simplified,” for easily determining the photopeak efficiency of NaI(Tl) scintillation detectors. These methods, along with a “general” solution method, were validated with calibrated radioactive sources such as 241Am, 57Co, 133Ba, 137Cs, and 60Co, and were used to commission a NaI(Tl) scintillation detector system at Chalk River Laboratories (CRL) for nuclear remediation. The system demonstrated high accuracy, making it suitable for screening unstable isotopes at contaminated sites. This screening tool significantly reduces the number of soil samples requiring detailed characterization, thereby lowering operational costs. The NaI(Tl) detector system was calibrated for near-contact geometry, which is commonly used at CRLs. Detection limits were established for this configuration. By improving the efficiency calibration of scintillation detectors, our study advances sustainable nuclear remediation practices, promoting environmental sustainability and cleaner production processes.

Multisite community-scale monitoring of respiratory and enteric viruses in the effluent of a nursing home and in the inlet of the local wastewater treatment plant.

The aim of this study was to evaluate whether community-level monitoring of respiratory and enteric viruses in wastewater can provide a comprehensive picture of local virus circulation. Wastewater samples were collected weekly at the wastewater treatment plant (WWTP) inlet and at the outlet of a nearby nursing home (NH) in Burgundy, France, during the winter period of 2022/2023. We searched for the pepper mild mottle virus as an indicator of fecal content as well as for the main respiratory viruses [severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), influenza, and respiratory syncytial virus] and enteric viruses (rotavirus, sapovirus, norovirus, astrovirus, and adenovirus). Samples were analyzed using real-time reverse transcription PCR-based methods. SARS-CoV-2 was the most frequently detected respiratory virus, with 66.7% of positive samples from the WWTP and 28.6% from the NH. Peaks of SARS-CoV-2 were consistent with the chronological incidence of infections recorded in the sentinel surveillance and the nearby hospital databases. The number of positive samples was lower in the NH than in WWTP for the three respiratory viruses. Enteric viruses were frequently detected, most often sapovirus and norovirus genogroup II, accounting both for 77.8% of positive samples in the WWTP and 57.1% and 37%, respectively, in the NH. The large circulation of sapovirus was unexpected in particular in the NH. Combined wastewater surveillance using simple optimized methods can be a valuable tool for monitoring viral circulation and may serve as a suitable early warning system for identifying both local outbreaks and the onset of epidemics. These results encourage the application of wastewater-based surveillance (WBS) to SARS-CoV2, norovirus, and sapovirus.IMPORTANCEWBS provides valuable information on the spread of epidemic viruses in the environment using appropriate and sensitive detection methods. By monitoring the circulation of viruses using reverse transcription PCR methods in wastewater from the inlet of a wastewater treatment plant and the outlet of a nearby retirement home (connected to the same collective sewer network), we aimed to demonstrate that implementing combined WBS at key community sites allows effective detection of the occurrence of respiratory (influenza, respiratory syncytial virus, and SARS-CoV-2) and enteric (norovirus, rotavirus, and sapovirus) virus infections within a given population. This analysis on a localized scale provided new information on the viral circulation in the two different sites. Implementing WBS to monitor the circulation or the emergence of infectious diseases is an important means of alerting the authorities and improving public health management. WBS could participate actively to the health of humans, animals, and the environment.

Application of a new highly multiplexed amplicon sequencing tool to evaluate Plasmodium falciparum antimalarial resistance and relatedness in individual and pooled samples from Dschang, Cameroon.

Resistance to antimalarial drugs remains a major obstacle to malaria elimination. Multiplexed, targeted amplicon sequencing is being adopted for surveilling resistance and dissecting the genetics of complex malaria infections. Moreover, genotyping of parasites and detection of molecular markers drug resistance in resource-limited regions requires open-source protocols for processing samples, using accessible reagents, and rapid methods for processing numerous samples including pooled sequencing. P lasmodium falciparum S treamlined M ultiplex A ntimalarial R esistance and R elatedness T esting ( Pf -SMARRT) is a PCR-based amplicon panel consisting of 15 amplicons targeting antimalarial resistance mutations and 9 amplicons targeting hypervariable regions. This assay uses oligonucleotide primers in two pools and a non-proprietary library and barcoding approach. We evaluated Pf -SMARRT using control mocked dried blood spots (DBS) at varying levels of parasitemia and a mixture of 3D7 and Dd2 strains at known frequencies, showing the ability to genotype at low parasite density and recall within-sample allele frequencies. We then piloted Pf -SMARRT to genotype 100 parasite isolates collected from uncomplicated malaria cases at three health facilities in Dschang, Western Cameroon. Antimalarial resistance genotyping showed high levels of sulfadoxine-pyrimethamine resistance mutations, including 31% prevalence of the DHPS A613S mutation. No K13 candidate or validated artemisinin partial resistance mutations were detected, but one low-level non-synonymous change was observed. Pf -SMARRT's hypervariable targets, used to assess complexity of infections and parasite diversity and relatedness, showed similar levels and patterns compared to molecular inversion probe (MIP) sequencing. While there was strong concordance of antimalarial resistance mutations between individual samples and pools, low-frequency variants in the pooled samples were often missed. Overall, Pf -SMARRT is a robust tool for assessing parasite relatedness and antimalarial drug resistance markers from both individual and pooled samples. Control samples support that accurate genotyping as low as 1 parasite per microliter is routinely possible. Malaria remains a critical global public health problem. Antimalarial drug resistance has repeatedly undermined control and the emergence of artemisinin partial resistance in Africa is the latest major challenge. Malaria molecular surveillance (MMS) has emerged as a powerful tool to monitor molecular markers of resistance and changes in the parasite population. Streamlined methods are needed that can be readily adopted in endemic countries. We developed P lasmodium falciparum S treamlined M ultiplex A ntimalarial R esistance and R elatedness T esting ( Pf -SMARRT), a multiplex amplicon deep sequencing approach that uses easily accessible products without proprietary steps and can be sequenced on any Illumina sequencer. We validated this tool using controls, including mocked dried blood spots, and then implemented it to evaluate resistance and parasite relatedness among 100 samples from Cameroon. The assay was able to reliably assess the within-sample allele frequency of antimalarial resistance markers and discriminate strains within and between individuals. We also evaluated a more cost-effective surveillance approach for antimalarial resistance polymorphisms using pooled samples. While within-pool frequencies of mutations were accurate in pools with higher numbers of samples, this resulted in the loss of the ability to detect variants uncommon in the pool. Overall Pf- SMARRT provides a new protocol for conducting MMS that is easily implementable in Africa.

Groundwater quality assessment for drinking and irrigation purposes in arid to semiarid region of Indus Basin of South Punjab, Pakistan

<p><a name="_Hlk173243332"><span style="background-color:white;color:black;font-family:"Times New Roman",serif;font-size:12.0pt;line-height:150%;">As the groundwater pollution is increasing, it is crucial to assess groundwater quality and characterize hydrogeochemistry accurately for long term water supply. </span></a><span style="background-color:white;color:black;font-family:"Times New Roman",serif;font-size:12.0pt;line-height:150%;">In this study groundwater samples were collected from 120 locations from Bahawalpur district and analyzed for electrical conductivity (EC), pH, total dissolved solids (TDS), sodium (Na<sup>+</sup>), potassium (K<sup>+</sup>), calcium (Ca<sup>2+</sup>), magnesium (Mg<sup>2+</sup>), carbonate (CO<sub>3</sub><sup>-</sup>), bicarbonate (HCO<sub>3</sub><sup>-</sup>), sulfate (SO<sub>4</sub><sup>2-</sup>), chloride (Cl<sup>-</sup>), nitrate (NO<sub>3</sub><sup>-</sup>), fluoride (F<sup>-</sup>) and heavy metals. The results obtained from the analysis of samples showed that the SAR of nearly 65% of samples collected from Bahawalpur district were in acceptable limit (<6 meq L<sup>-1</sup>), while 12% found to be unfit for irrigation (>10 meq L<sup>-1</sup>). Similarly, the RSC values depicted that 8% samples were unfit while 64% samples were falling in the category of fit irrigation water. Most of the water samples were predominantly Ca<sup>2+</sup>-HCO<sub>3</sub><sup>-</sup> and Na<sup>+</sup>-HCO<sub>3</sub><sup>-</sup> type, that were controlled by various processes of cation exchange, water-rock interaction, dissolution and evaporation. Some samples fall in the middle of diamond and lower triangles of Piper diagram are showing no dominant type of water (mixed water type) due to the complex influence of rock-water interactions as well as anthropogenic activities. The Water Quality Index (WQI) showed that out of total 120 sampling sites in Bahawalpur the number of excellent water samples were only 2% while 18% water samples were characterized as hazardous. Moreover, the number of samples falling in the categories of good, poor and very poor quality water were 36%, 29% and 13% respectively.</span></p>

Feature Bagging with Nested Rotations (FBNR) for anomaly detection in multivariate time series

Detecting anomalies in multivariate time series poses a significant challenge across various domains. The infrequent occurrence of anomalies in real-world data, as well as the lack of a large number of annotated samples, makes it a complex task for classification algorithms. Deep Neural Network approaches, based on Long Short-Term Memory (LSTMs), Autoencoders, and Variational Autoencoders (VAEs), among others, prove effective with handling imbalanced data. However, the same does not follow when such algorithms are applied on multivariate time-series, as their performance degrades significantly. Our main hypothesis is that the above is due to anomalies stemming from a small subset of the feature set. To mitigate the above issues in the multivariate setting, we propose forming an ensemble of base models by combining different feature selection and transformation techniques. The proposed processing pipeline includes applying a Feature Bagging techniques on multiple individual models, which considers separate feature subsets for each specific model. These subsets are then partitioned and transformed using multiple nested rotations derived from Principal Component Analysis (PCA). This approach aims to identify anomalies that arise from only a small portion of the feature set while also introduces diversity by transforming the subspaces. Each model provides an anomaly score, which are then aggregated, via an unsupervised decision fusion model. A semi-supervised fusion model was also explored, in which a Logistic Regressor was applied on the individual model outputs. The proposed methodology is evaluated on the Skoltech Anomaly Benchmark (SKAB), containing multivariate time series related to water flow in a closed circuit, as well as the Server Machine Dataset (SMD), which was collected from a large Internet company. The experimental results reveal that the proposed ensemble technique surpasses state-of-the-art algorithms. The unsupervised approach demonstrated a performance improvement of 2% for SKAB and 3% for SMD, compared to the baseline models. In the semi-supervised approach, the proposed method achieved a minimum of 10% improvement in terms of anomaly detection accuracy.

Innovative strategies for intelligent services in smart libraries in the information age based on linear discriminant analysis

With the advent of the information age, to provide better services and ensure the security management of libraries, intelligent facial recognition technology has gradually become a hot research direction in library management. Meanwhile, to further improve the comprehensive performance of facial recognition, this study attempts to integrate principal component analysis and linear discriminant analysis on the basis of analyzing the framework of recognition technology. Afterwards, it introduced support vector machines for recognition and classification, and proposed a new recognition model. The experimental results show that the recognition accuracy of the proposed model is up to 97 % in the ORL dataset and 94 % in the Yale dataset. The recognition error rate is as low as 0.1 % when the number of training samples is 215 and the number of iterations is 200. The model has the best recognition performance when the image size is 25 × 25 mm and the number of noises is 10. In addition, the model is particularly effective in recognition on single person color or gray images, with the highest P-value of 98.7 %, the highest R-value of 98.8 %, and the highest F1-value of 97.5 %. These results show that the proposed model significantly improves the accuracy and robustness of face recognition, and provides strong technical support for intelligent service innovation in smart libraries.

Investigating In Vivo Tumor Biomolecular Changes Following Radiation Therapy Using Raman Spectroscopy.

Treatment resistance is a major bottleneck in the success of cancer therapy. Early identification of the treatment response or lack thereof in patients can enable an earlier switch to alternative treatment strategies that can enhance response rates. Here, Raman spectroscopy was applied to monitor early tumor biomolecular changes in sensitive (UM-SCC-22B) and resistant (UM-SCC-47) head and neck tumor xenografts for the first time in in vivo murine tumor models in response to radiation therapy. We used a validated multivariate curve resolution-alternating least-squares (MCR-ALS) model to resolve complex multicomponent Raman spectra into individual pure spectra and their respective contributions. We observed a significant radiation-induced increase in the contributions of lipid-like species (p = 0.0291) in the radiation-sensitive UM-SCC-22B tumors at 48 h following radiation compared to the nonradiated baseline (prior to commencing treatment). We also observed an increase in the contribution of collagen-like species in the radiation-resistant UM-SCC-47 tumors at 24 h following radiation compared to the nonradiated baseline (p = 0.0125). In addition to the in vivo analysis, we performed ex vivo confocal Raman microscopic imaging of frozen sections derived from the same tumors. A comparison of all control and treated tumors revealed similar trends in the contributions of lipid-like and collagen-like species in both in vivo and ex vivo measurements; however, when evaluated as a function of time, longitudinal trends in the scores of collagen-like and lipid-like components were not consistent between the two data sets, likely due to sample numbers and differences in sampling depth at which information is obtained. Nevertheless, this study demonstrates the potential of fiber-based Raman spectroscopy for identifying early tumor microenvironmental changes in response to clinical doses of radiation therapy.

Natural gas leakage detection from offshore platform by OGI camera and unsupervised deep learning

Natural gas leak from offshore platform poses a potential to cause explosion disaster and bring significant causalities and economic losses. Existing deep learning-based leak detection approaches are limited by the requirement of a large number of labeled leak datasets, and also has worse performance in the complex and changeable marine environment. This study proposes a detection approach of natural gas leakage from offshore platform by integrating optical gas imaging (OGI) camera and unsupervised deep probability learning. In this approach, unsupervised deep learning is applied to learn the changeable infrared features of offshore platform, and variational Bayesian inference is integrated to provide the larger epistemic uncertainty contour corresponding to the infrared natural gas plume. An epistemic uncertainty-based detection score is proposed as the detection criterion to improve the accuracy of natural gas plume detection and localization. An OGI imaging experiment of natural gas leak from offshore platform is conducted to construct the benchmark dataset. With such datasets, two pre-defined parameters, namely Monte Carlo sampling number m = 100 and dropout probability p=0.1 are determined to guarantee the detection accuracy and efficiency. Comparison between the proposed approach and prevalent unsupervised deep learning approach is also conducted. The results demonstrate that our proposed approach has the higher detection accuracy with AUC = 0.9753, as well as the real-time capability with inference time of 4s/frame. Overall, our proposed approach provides a more accurate and generalized approach of natural gas detection for safety monitoring and detection management of offshore platforms.