Detection Scheme Research Articles

Kafka‐Shield: Kafka Streams‐based distributed detection scheme for IoT traffic‐based DDoS attacks

AbstractWith the rapid proliferation of insecure Internet of Things (IoT) devices, the security of Internet‐based applications and networks has become a prominent concern. One of the most significant security threats encountered in IoT environments is a Distributed Denial of Service (DDoS) attack. This attack can severely disrupt critical services and prevent smart devices from functioning normally, leading to severe consequences for businesses and individuals. It aims to overwhelm victims' resources, websites, and other services by flooding them with massive attack packets, making them inaccessible to legitimate users. Researchers have developed multiple detection schemes to detect DDoS attacks. As technology advances and other facilitating factors have increased, it is a challenge to identify such powerful attacks in real‐time. In this paper, we propose a novel distributed detection scheme for IoT network traffic‐based DDoS attacks by deploying it in a Kafka Streams processing framework named Kafka‐Shield. The Kafka‐Shield comprises two stages: design and deployment. Firstly, the detection scheme is designed on the Hadoop cluster employing a highly scalable H2O.ai machine learning platform. Secondly, a portable, scalable, and distributed detection scheme is deployed on the Kafka Streams processing framework. To analyze the incoming traffic data and categorize it into nine target classes in real time. Additionally, Kafka‐Shield stores each network flow with significant input features and the predicted outcome in the Hadoop Distributed File System (HDFS). It enables the development of new models or updating current ones. To validate the effectiveness of the Kafka‐Shield, we performed critical analysis using various configured attack scenarios. The experimental results affirm Kafka‐Shield's remarkable efficiency in detecting DDoS attacks. It has a detection rate of over 99% and can process 0.928 million traces in nearly 3.027 s.

Sensitive and on-Site Detection of Staphylococcus aureus Based on CRISPR/Cas 13a-Assisted Chemiluminescence Resonance Energy Transfer.

Developing a specific, sensitive, rapid, and on-site method for detecting pathogenic bacteria in food samples is critical to ensuring public safety. This article demonstrates a CRISPR/Cas13a system and a chemiluminescence resonance energy transfer (CRET) (CRISPR/Cas 13a-assisted CRET)-based strategy for sensitive and on-site detection of pathogenic bacteria in real samples. Once the hybrid double strand of aptamerS.aureus-cRNA recognizes the target model bacteria of Staphylococcus aureus (S. aureus), the released cRNA would bind with CRISPR/Cas 13a to form a complex of cRNA-CRISPR/Cas 13a, which could cleave the RNA molecule in the detecting probe of horseradish peroxidase (HRP) modified-gold nanoparticles (AuNPs) linked by RNA (AuNPs-RNA-HRP), resulting in an enhanced chemiluminescence signal due to the CRET "OFF" phenomenon after introducing the chemiluminescence substrate of luminol. The CRISPR/Cas 13a-assisted CRET strategy successfully detected S. aureus in drinking water and milk with detection limits of 20 and 30 cfu/mL, respectively, within the recovery of 90.07-105.50%. Furthermore, after integrating with an immunochromatographic test strip (ICTS), the CRISPR/Cas 13a-assisted CRET strategy achieved the on-site detection of as low as 102 cfu/mL of S. aureus in drinking water and milk via a smartphone, which is about 10 times lower than that in the previously reported AuNPs-based colorimetric ICTS, demonstrating a convenient and sensitive detection method for S. aureus in real samples.

Energy-Efficient and Highly Reliable Geographic Routing Based on Link Detection and Node Collaborative Scheduling in WSN

Energy efficiency and data reliability are important indicators to measure network performance in wireless sensor networks. In existing research schemes of routing protocols, the impact of node coverage on the network is often ignored, and the possibility that multiple sensor nodes may sense the same spatial point is not taken into account, which results in a waste of network resources, especially in large-scale networks. Apart from that, the blindness of geographic routing in data transmission has been troubling researchers, which means that the nodes are unable to determine the validity of data transmission. In order to solve the above problems, this paper innovatively combines the routing protocol with the coverage control technique and proposes the node collaborative scheduling algorithm, which fully considers the correlation characteristics between sensor nodes to reduce the number of active working nodes and the number of packets generated, to further reduce energy consumption and network delay and improve packet delivery rate. In order to solve the problem of unreliability of geographic routing, a highly reliable link detection and repair scheme is proposed to check the communication link status and repair the invalid link, which can greatly improve the packet delivery rate and throughput of the network, and has good robustness. A large number of experiments demonstrate the effectiveness and superiority of our proposed scheme and algorithm.

Diabetes and Cancer: A Twisted Bond

This paper presents an overview of the interconnection between various factors related to both cancer and type 2 diabetes mellitus (T2DM). Hyperglycemia, hyperinsulinemia, chronic inflammation, and obesity are involved in the development and progression of both diseases but, strong evidence for a direct causal relationship between diabetes and cancer, is lacking. Several studies described a relationship between hyperglycemia and cancer at the cellular, tissular and organismic levels but at the same time recent Mendelian randomization studies proved a significant causal relationship only between hyperglycemia and breast cancer. On the other hand, the association between both hyperinsulinemia and obesity and several cancer types appears to be robust as demonstrated by Mendelian randomized studies. Metabolic alterations, including the Warburg effect and excessive glucose consumption by tumors, are discussed, highlighting the potential impact of dietary restrictions, such as fasting and low-carb diets, on tumor growth and inflammation. Recent data indicates that circulating branched-chain amino acids levels, may represent novel biomarkers that may contribute to both better diabetes control and early pancreatic cancer detection. Understanding the underlying mechanisms and shared risk factors between cancer and T2DM can provide valuable insights for cancer prevention, early detection, and management strategies.

Polarization detection for cylindrical vector beams empowered by pixelated metasurfaces

Abstract Polarimetry plays an indispensable role in the light-matter interactions. Nevertheless, conventional components developed for polarization measurements suffer from bulky volume and spatial alignment schemes, causing them to reveal limited performance in determining inhomogeneous polarization distributions. Here, we propose a polarization detection scheme based on pixelated all-dielectric metasurfaces using spin-multiplexing coding techniques. The polarization resolving capability of the pixelated metasurface under homogeneous linearly polarized illumination was first evaluated, and the extracted peak coordinates were used to establish an exact functional relationship with the azimuthal angle. Subsequently, the measurement of spatial inhomogeneous polarization was further explored with a focusing matrix assembled from pixelated metasurfaces. The proposed polarization detection strategy can be extended to other spectral bands without discrimination, stimulating potential applications in high-resolution imaging, sensing and data communication.

Toehold Strand Displacement-Mediated Exponential HCR for Highly Sensitive and Specific Analysis of miRNA in Living Cells.

As an important disease biomarker, the development of sensitive detection strategies for miRNA, especially intracellular miRNA imaging strategies, is helpful for early diagnosis of diseases, pathological research, and drug development. Hybridization chain reaction (HCR) is widely used for miRNA imaging analysis because of its high specificity and lack of biological enzymes. However, the classic HCR reaction exhibits linear amplification with low efficiency, limiting its use for the rapid analysis of trace miRNA in living cells. To address this problem, we proposed a toehold-mediated exponential HCR (TEHCR) to achieve highly sensitive and efficient imaging of miRNA in living cells using β-FeOOH nanoparticles as transfection vectors. The detection limit of TEHCR was as low as 92.7 fM, which was 8.8 × 103 times lower compared to traditional HCR, and it can effectively distinguish single-base mismatch with high specificity. The TEHCR can also effectively distinguish the different expression levels of miRNA in cancer cells and normal cells. Furthermore, TEHCR can be used to construct OR logic gates for dual miRNA analysis without the need for additional probes, demonstrating high flexibility. This method is expected to play an important role in clinical miRNA-related disease diagnosis and drug development as well as to promote the development of logic gates.

Parents and Children: Distinguishing Multimodal DeepFakes from Natural Images

Recent advancements in diffusion models have enabled the generation of realistic deepfakes from textual prompts in natural language. While these models have numerous benefits across various sectors, they have also raised concerns about the potential misuse of fake images and cast new pressures on fake image detection. In this work, we pioneer a systematic study on deepfake detection generated by state-of-the-art diffusion models. Firstly, we conduct a comprehensive analysis of the performance of contrastive and classification-based visual features, respectively extracted from CLIP-based models and ResNet or ViT-based architectures trained on image classification datasets. Our results demonstrate that fake images share common low-level cues, which render them easily recognizable. Further, we devise a multimodal setting wherein fake images are synthesized by different textual captions, which are used as seeds for a generator. Under this setting, we quantify the performance of fake detection strategies and introduce a contrastive-based disentangling method that lets us analyze the role of the semantics of textual descriptions and low-level perceptual cues. Finally, we release a new dataset, called COCOFake, containing about 1.2M images generated from the original COCO image-caption pairs using two recent text-to-image diffusion models, namely Stable Diffusion v1.4 and v2.0.

Cyber Security and Digital Economy: Opportunities, Growth and Challenges

The digital age has witnessed the exponential growth of the international economy, propelled by digital technology and the deepening interconnectedness of the global landscape. However, amidst this rapid expansion lies a plethora of Cyber Security (CS) risks within the Digital Economy (DE), ranging from hacker assaults to data breaches and intellectual property theft. The heightened importance of CS resonates across organizations, nations, and individuals worldwide. This study delves into the dynamics of development and expansion within the DE while concurrently addressing the looming threats to CS. Central to its analysis is the imperative of implementing efficient CS protocols, encompassing preventive measures, detection strategies, and swift response mechanisms against online threats. Moreover, the discourse extends to strategies for enhancing CS resilience through investments in infrastructure, capacity building initiatives, and public awareness campaigns, involving both corporate entities and governmental bodies. Furthermore, this research probes into the intricate interplay between cutting-edge technologies such as blockchain, Artificial Intelligence (AI), and the Internet of Things (IoT) and their impact on CS. While these technologies hold immense potential to revolutionize the DE landscape, they concurrently introduce novel security vulnerabilities that malicious actors may exploit. In conclusion, the study underscores that while the DE offers boundless opportunities for development and innovation, robust CS measures stand as indispensable safeguards against online threats, ensuring the sustained viability and integrity of the sector in the long term.

SEA Anemone-inspired Conducting Polymer Sensing Platform for Integrated Detection of Tumor Protein Marker and Circulating Tumor Cell.

Combining the detection of tumor protein markers with the capture of circulating tumor cells (CTCs) represents an ultra-promising approach for early tumor detection. However, current methodologies have not yet achieved the necessary low detection limits and efficient capture. Here, we introduced a novel polypyrrole nanotentacles sensing platform featuring anemone-like structures capable of simultaneously detecting protein biomarkers and capturing CTCs. The incorporation of nanotentacles significantly enhanced the electrode surface area, providing abundant active sites for antibody binding. This enhancement allowed detecting nucleus matrix protein22 (NMP22) and bladder tumor antigen (BTA) with 2.39 and 3.12 pg/mL detection limit, respectively. Furthermore, our developed sensing platform effectively captured MCF-7 cells in blood samples with a detection limit of fewer than 10 cells/mL, attributed to the synergistic multivalent binding facilitated by the specific recognition antibodies and the positive charge on the nanotentacles surface. This sensing platform demonstrated excellent detection capabilities and outstanding capture efficiency, offering a simple, accurate, and efficient strategy for early tumor detection. This article is protected by copyright. All rights reserved.

Predictive modelling and identification of key risk factors for stroke using machine learning

Strokes are a leading global cause of mortality, underscoring the need for early detection and prevention strategies. However, addressing hidden risk factors and achieving accurate prediction become particularly challenging in the presence of imbalanced and missing data. This study encompasses three imputation techniques to deal with missing data. To tackle data imbalance, it employs the synthetic minority oversampling technique (SMOTE). The study initiates with a baseline model and subsequently employs an extensive range of advanced models. This study thoroughly evaluates the performance of these models by employing k-fold cross-validation on various imbalanced and balanced datasets. The findings reveal that age, body mass index (BMI), average glucose level, heart disease, hypertension, and marital status are the most influential features in predicting strokes. Furthermore, a Dense Stacking Ensemble (DSE) model is built upon previous advanced models after fine-tuning, with the best-performing model as a meta-classifier. The DSE model demonstrated over 96% accuracy across diverse datasets, with an AUC score of 83.94% on imbalanced imputed dataset and 98.92% on balanced one. This research underscores the remarkable performance of the DSE model, compared to the previous research on the same dataset. It highlights the model's potential for early stroke detection to improve patient outcomes.

Absorption by a Layered Microbolometer Pixel’s Active Element

Microbolometer arrays, i.e., arrays of micro-scale pixels sensing temperature via resistance changes, have proven to be an effective basis for real-time imaging instrumentation in infrared as well as terahertz frequencies. In previous work, a design of THz and IR absorbing nano-laminates of dielectric and metal layers was studied. It was shown via numerical modeling that absorption may be maximized by appropriate choices of thickness, permittivity and conductivity. In this work, an analytical approach to the problem is formulated based on the standard recursive multiple reflection formulas for multi-layered planar structures. The results fully confirm and extend previous numerical work. A previous relationship between wavelength and silicon thickness for maximum absorption, derived numerically for specific parameter combinations, is now generalized in a parametric closed form. The method can be extended to include multiple lossy dielectric layers and may serve as a tool for optimizing the absorption characteristics of more complex layered absorbing structures. This could enhance the sensitivity of the detection scheme of interest, providing benefits in terms of cost, efficiency, precision, and adjustability.

Response to editorial: Gastric and pancreatic cancer trends in Spain.

This response to an editorial discusses recent trends in gastric and pancreatic cancer mortality in Spain. The authors acknowledge the contrasting trajectories - a decrease in gastric cancer and an increase in pancreatic cancer - and attribute them to differing causes. Public health measures, particularly H. pylori eradication, are credited for the decline in gastric cancer. The authors also present unpublished data showing a worrying rise in gastric cancer diagnoses among young men. They emphasize the need for continued monitoring, early detection strategies, and preventative measures to tackle both cancers, particularly focusing on pancreatic cancer research due to its rising mortality.

Edge server enhanced secure and privacy preserving federated learning

Federated learning (FL) has been smoothly embedded into current IoT edge computing architecture, and hatching advanced IoT-FL applications. While so, sensitive messages generated on terminal sides brings privacy issues; and vulnerable terminal devices are easy targets for Byzantine adversaries, they inject malicious data to manipulate the IoT-FL system. Now an inherent dilemma is, the privacy preserving requires anonymous indistinguishable individual characteristics, while the malicious adversary detection requires transparent distinguishable results. Existing fragmented schemes are not able to handle both problems coherently. This time, inherit our previous study, based on the IoT edge computing architecture, we utilize Homomorphic Encryption (HE) and Threshold Secret Sharing (SS) methods, propose a joint scheme (namely, Sec-IoTFL) for anonymous adversary detection. Specifically, we batch encode the clients’ training result vectors (message) as polynomials, and split them as secret pieces with a unified SS key set; then edge server and cloud server joint deal with these secret pieces in a specialized flow, where Byzantine adversaries will be filtered out. The theoretical analysis and solid experimental results suggest, in our scheme, local sensitive data is well preserved, and malicious behavior is precisely screened. Comparing with traditional methods, our Sec-IoTFL scheme shows the superiority in accuracy and efficiency.

Recyclable Au@R–Fe3O4/g-C3N4 substrates for rapid SERS detection and degradation of multiple pollutants

Developing a Surface-enhanced Raman spectroscopy (SERS) method with excellent detecting ability, good recyclability and analyzing multiple pollutants rapidly are critical for evaluation of water quality in emergency pollution affairs. While constructing a multifunctional substrate with these characteristics to realize the application of SERS in water quality monitoring remains a challenge. In this work, a reusable Au@R–Fe3O4/g-C3N4 SERS substrate is prepared by loading Au nanoparticles (Au NPs) on Fe3O4 nanorings (R–Fe3O4) and the formed Au@R–Fe3O4 is further combined with g-C3N4 nanosheets through a simple electrostatic assembly method. The Au@R–Fe3O4/g-C3N4 nanocomposite presents multifunction of magnetic enrichment, SERS signal enhancement, multiple pollutants analyzing, and photocatalytic activity, which achieves quantitative detection of rhodamine B (RhB), tetracycline hydrochloride (TC), and 4-chlorophenol (4-CP), with detection limits of 5.30 × 10−9, 7.50 × 10−8, 7.69 × 10−8 mol/L, respectively. Furthermore, the recyclable detection capability of Au@R–Fe3O4/g-C3N4 for multi components is demonstrated by the strong SERS signal after 9 cycles of “detection-degradation” processes. Combined with good uniformity and stability, this SERS method based on Au@R–Fe3O4/g-C3N4 substrate provides a new strategy for the multi-pollutants detection and degradation in water environment.

AI-enhanced integration of genetic and medical imaging data for risk assessment of Type 2 diabetes.

Type 2 diabetes (T2D) presents a formidable global health challenge, highlighted by its escalating prevalence, underscoring the critical need for precision health strategies and early detection initiatives. Leveraging artificial intelligence, particularly eXtreme Gradient Boosting (XGBoost), we devise robust risk assessment models for T2D. Drawing upon comprehensive genetic and medical imaging datasets from 68,911 individuals in the Taiwan Biobank, our models integrate Polygenic Risk Scores (PRS), Multi-image Risk Scores (MRS), and demographic variables, such as age, sex, and T2D family history. Here, we show that our model achieves an Area Under the Receiver Operating Curve (AUC) of 0.94, effectively identifying high-risk T2D subgroups. A streamlined model featuring eight key variables also maintains a high AUC of 0.939. This high accuracy for T2D risk assessment promises to catalyze early detection and preventive strategies. Moreover, we introduce an accessible online risk assessment tool for T2D, facilitating broader applicability and dissemination of our findings.

Eutrophication and HAB Occurrence Control in Lakes of Different Origins: A Multi-Source Remote Sensing Detection Strategy

Remote sensing techniques have become pivotal in monitoring algal blooms and population dynamics in freshwater bodies, particularly to assess the ecological risks associated with eutrophication. This study focuses on remote sensing methods for the analysis of 4 Italian lakes with diverse geological origins, leveraging water quality samples and data from the Sentinel-2 and Landsat 5.7–8 platforms. Chl-a, a well-correlated indicator of phytoplankton biomass abundance and eutrophication, was estimated using ordinary least squares linear regression to calibrate surface reflectance with chl-a concentrations. Temporal gaps between sample and image acquisition were considered, and atmospheric correction dedicated to water surfaces was implemented using ACOLITE and those specific to each satellite platform. The developed models achieved determination coefficients higher than 0.69 with mean square errors close to 3 mg/m3 for water bodies with low turbidity. Furthermore, the time series described by the models portray the seasonal variations in the lakes water bodies.

Attack detection scheme for continuous variable quantum key distribution based on a graph neural network

The practical security of a continuous-variable quantum key distribution (CV-QKD) system is vulnerable to various attack strategies due to the significant difference between the idealized theoretical model and the practical physical system. The existing countermeasures against these attacks involve exploiting different real-time monitoring modules, which presents a challenge in effectively classifying attacks. We investigate a graph neural network (GNN)-based attack detection scheme for CV-QKD, which models data as a graph structure using three different methods for various conditions. Particularly, one of the proposed methods requires no additional devices and can detect attacks with over 99% accuracy. The algorithm can be expanded to different scenarios without additional training and can achieve a detection efficiency of more than 95%. Furthermore, our proposed scheme incorporates anomaly detection algorithms into the detection module, enabling 85% effective detection of partially unknown attacks with minimal security data.

Building Resilience in Banking Against Fraud with Hyper Ensemble Machine Learning and Anomaly Detection Strategies

Building Resilience in Banking Against Fraud with Hyper Ensemble Machine Learning and Anomaly Detection Strategies

Disease burden and high-risk populations for complications in patients with acute respiratory infections: a scoping review

BackgroundAcute respiratory infections (ARIs) represent a significant public health concern in the U.S. This study aimed to describe the disease burden of ARIs and identify U.S. populations at high risk of developing complications.MethodsThis scoping review searched PubMed and EBSCO databases to analyze U.S. studies from 2013 to 2022, focusing on disease burden, complications, and high-risk populations associated with ARIs.ResultsThe study included 60 studies and showed that ARI is associated with a significant disease burden and healthcare resource utilization (HRU). In 2019, respiratory infection and tuberculosis caused 339,703 cases per 100,000 people, with most cases being upper respiratory infections and most deaths being lower respiratory infections. ARI is responsible for millions of outpatient visits, especially for influenza and pneumococcal pneumonia, and indirect costs of billions of dollars. ARI is caused by multiple pathogens and poses a significant burden on hospitalizations and outpatient visits. Risk factors for HRU associated with ARI include age, chronic conditions, and socioeconomic factors.ConclusionThe review underscores the substantial disease burden of ARIs and the influence of age, chronic conditions, and socioeconomic status on developing complications. It highlights the necessity for targeted strategies for high-risk populations and effective pathogen detection to prevent severe complications and reduce HRU.

Electrochemiluminescence in Graphitic Carbon Nitride Decorated with Silver Nanoparticles for Dopamine Determination Using Machine Learning.

Electrochemiluminescence (ECL) luminophores with wavelength-tunable multicolor emissions are essential for multicolor ECL imaging detection and multiplexed analysis. In this work, silver nanoparticle (Ag NP)-decorated graphitic carbon nitride (g-CN@Ag) nanocomposites were synthesized. The morphology, chemical composition, structure, and ECL property of g-CN@Ag were investigated. The prepared g-CN, g-CN@Ag1, g-CN@Ag5, and g-CN@Ag10 can produce blue, blue-green, chartreuse, and yellow colored ECL emissions, respectively, by using K2S2O8 as the coreagent. The ECL emission wavelength of g-CN@Ag can be regulated from 460 to 565 nm by modulating the content of the immobilized Ag NPs. Then, a multicolor ECL detection array was fabricated by using g-CN, g-CN@Ag1, g-CN@Ag5, and g-CN@Ag10 as four ECL luminophores. Dopamine was detected based on its inhibition effect on the multicolor ECL emissions. The linear range is from 0.1 nM to 1 mM with the lowest detection limit of 44 pM. Then, machine learning-assisted multiparameter concentration prediction of dopamine was further carried out by combining the deep neural network (DNN) algorithm. This work provides a new avenue to regulate the ECL emission wavelength of g-CN by using the metal nanoparticle modification strategy and presents an effective machine learning-assisted multicolor ECL detection strategy for accurate multiparameter quantitative detection.