Current Detection Methods Research Articles

DeepWhaleNet: Climate Change-Aware FFT-Based Deep Neural Network for Passive Acoustic Monitoring

Climate change poses severe risks to the survival of many whale populations, whose habitats and migration patterns are affected by environmental changes. To detect these whales effectively, especially in deep-sea environments, we need to use AI-based techniques to handle the acoustic diversity and variability of different species. However, current methods for whale detection are based on pre- and post-processing steps that reduce their efficiency and generalizability. To address this issue, we present DeepWhaleNet, a novel deep-learning model that automates whale detection in Underwater Passive Acoustic Monitoring datasets. DeepWhaleNet simplifies the detection process by extracting relevant features from raw log-power spectrograms and helps protect these threatened species by supporting conservation efforts. Our model uses a larger short-time Fourier transform as input and a custom ResNet-18 architecture for classification, which enables it to separate whale sounds from noise and capture their temporal and spectral characteristics. We evaluate the performance of DeepWhaleNet and show that it surpasses state-of-the-art methods, achieving an 8.3% improvement in the F-1 score and 21% higher average precision of binary relevance than the baseline method. Moreover, our model demonstrates its versatility and suitability for species-specific retrieval problems through an ablation study on multi-label retrieval problems and a 99.1% recall for Blue Whales.

I-MPaFS: enhancing EDoS attack detection in cloud computing through a data-driven approach

Cloud computing offers cost-effective IT solutions but is susceptible to security threats, particularly the Economic Denial of Sustainability (EDoS) attack. EDoS exploits cloud elasticity and the pay-per-use billing model, forcing users to incur unnecessary costs. This research introduces the Integrated Model Prediction and Feature Selection (I-MPaFS) framework to address EDoS attacks. I-MPaFS framework enhances an existing dataset to improve performance, using the generated data to build a Random Forest model for EDoS detection. Our investigation employs the UNSW-NB15, CSE-CIC-IDS18 and NSL-KDD datasets, demonstrating the proposed method’s superiority over existing techniques. The model achieved recall scores of 99.45% on the UNSW-NB15 dataset, 98.19% on the CSE-CIC-IDS18 dataset, and 99.82% on the NSL-KDD dataset, highlighting its reliability and efficacy in safeguarding cloud users from financial exploitation. This study contributes to the field by evaluating current EDoS detection methods, introducing the I-MPaFS framework, validating its performance with benchmark datasets, and comparing its effectiveness against state-of-the-art techniques. The findings affirm the significant potential of I-MPaFS in enhancing cloud security and protecting users from EDoS attacks.

MonoCAPE: Monocular 3D object detection with coordinate-aware position embeddings

3D monocular detection remains to be a focal point of research, particularly due to its capacity to deliver available precision under conditions of low cost and simplified configurations, making it especially valuable in fields like autonomous driving. Current 3D object detection methods often overlook the spatial information missing from images, which is critical to spatial perception, and optimize bounding box attributes separately, failing to meet the requirements of autonomous driving. We introduce MonoCAPE, a novel 3D detection framework addressing these issues by encoding spatial information and co-optimizing attributes through a Coordinate-Aware Position Encoding (CAPE) Generator and a Task Co-optimization Strategy (TCS). The CAPE Generator produces sparse positional embeddings, enabling spatial awareness with low computational cost, while the TCS utilizes Gaussian modeling to prevent suboptimal outputs. In this way, our framework comprehensively takes into account what existing approaches ignore. Extensive experiments on the KITTI dataset demonstrate MonoCAPE significantly improves AP3D and APBEV metrics compared to existing advanced methods.

Utilizing graph neural networks for adverse health detection and personalized decision making in sensor-based remote monitoring for dementia care

Background:Sensor-based remote health monitoring is increasingly used to detect adverse health in people living with dementia (PLwD) at home, aiming to prevent hospitalizations and reduce caregiver burden. However, home sensor data is often noisy, overly granular, and suffers from unreliable labeling, data drift and high variability between households. Current anomaly detection methods lack generalizability and personalization, often requiring anomaly-free training data and frequent model updates. Objective:To develop a lightweight, explainable, self-supervised approach with personalized alert thresholds to detect adverse health events in PLwD, using changes in home activity. Methods:We hypothesized that health downturns manifest as detectable shifts in household movement patterns. Our approach leverages a Graph Barlow Twins contrastive model, which uses granular activity data and a macroscopic view to extract noise-robust, high-level and low-level discriminative features that represent daily activity patterns. Household-personalized alert thresholds are calculated based on clinician-set target alert rates, and daily anomaly scores are compared against these thresholds, triggering alerts for the clinical monitoring team. Model attention weights support explainability. Data were collected from a real-world dataset by the UK Dementia Research Institute (August 2019-April 2022). Results:Our model outperformed state-of-the-art temporal graph algorithms in detecting agitation and fall events across three patient cohorts, achieving 81% average recall and 88% generalizability at a target alert rate of 7%. Conclusion:We developed a novel, lightweight, explainable, and personalized Graph Barlow Twins model for real-world remote health monitoring in dementia care, with potential for broader applications in healthcare and sensor-based environments.

A novel machine-learning aided platform for rapid detection of urine ESBLs and carbapenemases: URECA-LAMP.

Pathogenic gram-negative bacteria frequently carry genes encoding extended-spectrum beta-lactamases (ESBL) and/or carbapenemases. Of great concern are carbapenem resistant Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa, and Acinetobacter baumannii. Despite the need for rapid AMR diagnostics globally, current molecular detection methods often require expensive equipment and trained personnel. Here, we present a novel machine-learning-aided platform for the rapid detection of ESBLs and carbapenemases using Loop-mediated isothermal Amplification (LAMP). The platform consists of (i) an affordable device for sample lysis, LAMP amplification, and visual fluorometric detection; (ii) a LAMP screening panel to detect the most common ESBL and carbapenemase genes; and (iii) a smartphone application for automated interpretation of results. Validation studies on clinical isolates and urine samples demonstrated percent positive and negative agreements above 95% for all targets. Accuracy, precision, and recall values of the machine learning model deployed in the smartphone application were all above 92%. Providing a simplified workflow, minimal operation training, and results in less than an hour, this study demonstrated the platform's feasibility for near-patient testing in resource-limited settings.IMPORTANCEExtended-spectrum beta-lactamases (ESBL) and carbapenemases confer resistance to third-generation cephalosporins and carbapenems in pathogenic Gram-negative bacteria such as Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa, and Acinetobacter baumannii. Conventional antimicrobial susceptibility testing is based on phenotypic methods, and results can take several days to be obtained. Current genotypic detection methods can be rapid but require expensive equipment and trained personnel. In this study, we present a novel machine learning-aided platform for the rapid detection of ESBLs and carbapenemases using Loop-mediated isothermal Amplification (LAMP). The validation of the platform demonstrated percent positive and negative agreements above 95% for all targets. The newly developed platform provided a simplified workflow, minimal technical training, and results in less than an hour. This study demonstrated the platform's feasibility for rapid testing of ESBL and carbapenemases in bacteria and urine specimens.

Helmet Wearing Detection Algorithm Based on YOLOv5s-FCW

An enhanced algorithm, YOLOv5s-FCW, is put forward in this study to tackle the problems that exist in the current helmet detection (HD) methods. These issues include having too many parameters, a complex network, and large computation requirements, making it unsuitable for deployment on embedded and other devices. Additionally, existing algorithms struggle with detecting small targets and do not achieve high enough recognition accuracy. Firstly, the YOLOv5s backbone network is replaced by FasterNet for feature extraction (FE), which reduces the number of parameters and computational effort in the network. Secondly, a convolutional block attention module (CBAM) is added to the YOLOv5 model to improve the detection model’s ability to detect small objects such as helmets by increasing its attention to them. Finally, to enhance model convergence, the WIoU_Loss loss function is adopted instead of the GIoU_Loss loss function. As reported by the experimental results, the YOLOv5s-FCW algorithm proposed in this study has improved accuracy by 4.6% compared to the baseline algorithm. The proposed approach not only enhances detection concerning small and obscured targets but also reduces computation for the YOLOv5s model by 20%, thereby decreasing the hardware cost while maintaining a higher average accuracy regarding detection.

A DNA Polymerase Variant Senses the Epigenetic Marker 5-Methylcytosine by Increased Misincorporation.

Dysregulation of DNA methylation is associated with human disease, particularly cancer, and the assessment of aberrant methylation patterns holds great promise for clinical diagnostics. However, DNA polymerases do not effectively discriminate between processing 5-methylcytosine (5 mC) and unmethylated cytosine, resulting in the silencing of methylation information during amplification or sequencing. As a result, current detection methods require multi-step DNA conversion treatments or careful analysis of sequencing data to decipher individual 5 mC bases. To overcome these challenges, we propose a novel DNA polymerase-mediated 5 mC detection approach. Here, we describe the engineering of a thermostable DNA polymerase variant derived from Thermus aquaticus with altered fidelity towards 5 mC. Using a screening-based evolutionary approach, we have identified a DNA polymerase that exhibits increased misincorporation towards 5 mC during DNA synthesis. This DNA polymerase generates mutation signatures at methylated CpG sites, allowing direct detection of 5 mC by reading an increased error rate after sequencing without prior treatment of the sample DNA.

Enzyme Activity-Based Optical Probe: Imaging Aminopeptidases N in Vulnerable Plaque and Serum.

Cardiovascular disease, a chronic and progressive arterial wall disease, is increasingly recognized for its clinical significance. Aminopeptidases N (APN), crucial in the pathophysiological processes of vulnerable plaque, have been linked to endothelial dysfunction, oxidative stress, and plaque formation, thus highlighting their potential as biomarkers for disease progression. However, current detection methods for APN in body fluids and in vivo have limitations, including insufficient sensitivity and specificity, time delays, and the inability to directly reflect enzyme activity in plaques. To address these challenges, we developed an optical probe, HD-APN, for in vivo imaging of aminopeptidases, providing a potential implementation in cardiovascular disease. Our work demonstrated the applicability of HD-APN for specific monitoring of aminopeptidase levels in plaques and serum, shedding light on its potential for further research in cardiovascular disease.

A novel double-door-opening sensor for visual determination of cardiac troponin I based on DNA hydrogel and bimetallic nanozyme

Cardiac troponin I (cTnI), as a cardiac biomarker, holds significant importance in the diagnosis of acute myocardial infarction. However, the current detection methods mostly require specialized personnel and large analytical instruments, making it difficult to achieve convenient and on-site testing. This situation leads to delayed disease diagnosis and treatment, that increases patient suffering, and reduces the cure rate. This strategy presents the development of a portable visual DNA hydrogel colorimetric sensing platform based on a smartphone. Utilizing dual-mode detection with ultraviolet signals and solution colorimetry, the platform achieves ultra-sensitive and real-time detection of cTnI. Furthermore, optimization of detection conditions, such as the amount of polyacrylamide, reaction time, aptamer concentration, encapsulation of the nanozyme, incubation time, and reaction temperature, were performed based on this platform. The proposed ultraviolet and visual detection platforms exhibited good linear relationships with the signal within the ranges of 0.003 ng mL−1 to 10.00 ng mL−1 and 0.01 ng mL−1 to 7.00 ng mL−1, with detection limits of 2.57 pg mL−1 and 0.013 pg mL−1, respectively. Additionally, utilizing 3D printing technology, a portable detection device was designed and employed for the detection of cTnI concentrations in human serum samples. Whatever in the initial and spiked samples, the results showed high sensitivities. The sensitivity and convenience of the sensor in detecting cTnI make it promising for home testing of patients, with broad market prospects.

NTLFlowLyzer: Towards generating an intrusion detection dataset and intruders behavior profiling through network and transport layers traffic analysis and pattern extraction

Network security remains a critical concern in modern computing systems due to the constant emergence of threats and attacks. This paper introduces a comprehensive behavioral profiling solution to address the limitations of current intrusion detection methods in identifying zero-day attacks and novel malicious behaviors. Beginning with raw network data, the proposed framework progresses through multiple stages, ultimately culminating in the creation of activity-specific profiles. Central to this approach is NTLFlowLyzer, a novel network traffic analyzer, which generates an updated dataset, BCCC-CIC-IDS2017, for enhanced profile generation. The core of the profiling system leverages the distinct behaviors exhibited by individual features and the diverse correlations observed across various activities. The profiling procedure attains accuracy and robustness by integrating a novel feature selection algorithm and a pattern extraction process. Furthermore, behavior similarity is introduced to quantify the resemblance between activities based on their features and behaviors. We rigorously evaluate the effectiveness of our model by subjecting it to comprehensive testing, followed by meticulous comparison with previous works. Our proposed framework proficiently characterizes eight malicious activities with an accuracy rate surpassing 99.8%, while displaying promising performance in profiling various other activities. These findings, derived from our comprehensive experiments, provide valuable guidance for accurately implementing behavioral profiling.

Ship detection method based on attention guidance and multi-sample decision making

Single-stage target detection methods have the characteristics of fast training speed and short detection time. However, its feature pyramid network is difficult to suppress the background and noise information of SAR ship images, and the detection head has prediction errors. To address this problem, this paper proposes a detection model based on attention guidance and multi-sample decision for synthetic aperture radar ship detection. Firstly, an attention guidance network is proposed and added to the highest level of the feature pyramid to suppress background and noise interference, thereby improving the representation ability of features. Secondly, a multi-sample decision network is proposed to participate in the prediction of target position. This network alleviates the impact of prediction errors on detection results by increasing the number of samples output in the regression branch. Finally, a novel maximum likelihood loss function is designed. This loss function constructs a maximum likelihood function using the samples output from the multi-sample decision network, which is used to standardize the training of the decision network and further improve the accuracy of target positioning. Taking the RetinaNet network model as the baseline method, compared with the baseline method and the current advanced target detection methods, this method shows the highest detection accuracy on the ship detection dataset SSDD, with AP reaching 52.8%. Compared with the baseline method, the proposed method improves the AP evaluation index 3.4% ∼ 5.7%, and the training parameter Params only increases by 2.03 M, and the frame rate FPS only decreases 0.5Iter/s.

Comparative study of pathogen detection methods for central nervous system infections: laboratory testing of tuberculous meningitis

BackgroundTuberculous meningitis (TBM) is a severe central nervous system (CNS) infection with a challenging diagnosis due to inadequate detection methods. This study evaluated current clinical detection methods and their applicability.MethodsA cohort of 514 CNS infection patients from 2018 to 2020 was studied. Data on general demographics, Cerebrospinal Fluid (CSF) analysis, epidemiology, and clinical outcomes were collected. TBM patients were identified, and the sensitivities of mmetagenomic next-generation sequencing (NGS), GeneXpert, and microbial culture were compared. Kappa statistic assessed the consistency between methods.ResultsAmong the patients involved, TBM (29%) and neurosyphilis (25%) were the two most prevalent CNS infections. CSF analysis indicated that 76% of patients had leukocytosis, suggesting a potential CNS inflammation. In TBM cases, 92.5% had elevated CSF protein and leukocyte counts. Moreover, the percentage of positive mNGS results was 55.6%. GeneXpert and MTB cultures alone had lower sensitivity, but combined use resulted in a 53.4% positive rate.ConclusionsThis study highlights the high sensitivity of mNGS, comparable to GeneXpert and MTB culture. The combined methods are cost-effective and straightforward, and can partially substitute for mNGS, offering valuable alternatives for TBM diagnosis and providing insights into multiple diagnostic strategies in clinical practice.

Seismic fault detection with sliding windowed differential cepstrum–based coherence analysis

AbstractCepstral decomposition is beneficial for highlighting certain geological features within the particular quefrency bands which may be deeply buried within the wide quefrency range of the seismic data. Converting seismic traces into the corresponding cepstrum components can better analyse some characteristics of underground strata than the traditional spectral decomposition methods. We propose the sliding windowed differential cepstrum–based coherence analysis approach to delineate the fault features. First, the data are decomposed using a sliding windowed differential cepstrum, which results in multi‐cepstrum data of corresponding quefrency of certain bandwidth. These different multi‐cepstrum data may highlight the different stratigraphic features in a certain quefrency band. We select the first‐order common quefrency volume as the featured attribute. Then, eigenstructure‐based coherence is applied on the first‐order common quefrency data volume to statistically obtain the fault detection result with a finer and sharper image. Synthetic data and field data examples show that the proposed method has the ability to better visualize all the possible subtle and minor faults present in the data more accurately and discernibly than the traditional coherence method. Compared with the ant‐tracking method, the proposed method is more effective in revealing the major faults. It is hoped that this work will complement current fault detection methods with the addition of the cepstral‐based method.

CN-ViT- Visual Object Detection with VisionTransformer

Object detection has always been an important and challenging task in the field of computer vision. In recent years, Vision Transformers (ViT) have achieved remarkable results on image classification tasks, demonstrating its potential in vision tasks. In this paper, we propose CN-ViT, a novel Vision Transformer based visual object detection model.CN-ViT effectively improves the accuracy and robustness of object detection by combining the advantages of self-attention mechanism and convolutional neural network, and introducing the GCCA (Global Context Block and Coordinate Attention) module. In this paper, CN-ViT is evaluated on the Mini COCO standard dataset. The experimental results suggest that CN-ViT may outperform current mainstream object detection methods in terms of detection accuracy and speed. This study sheds light on the potential of Transformer architectures for complex visual tasks and offers valuable insights for future research in this area.

A Microbial Cocaine Bioreporter

The continuous emergence of new illegal compounds, particularly psychoactive chemicals, poses significant challenges for current drug detection methods. Developing new protocols and kits for each new drug requires substantial time, effort, and dedicated manpower. Whole-cell bacterial bioreporters have been proven capable of detecting diverse hazardous compounds in both laboratory and field settings, identifying not only single compounds but also chemical families. We present the development of a microbial bioreporter for the detection of cocaine, the nervous system stimulant that is the second-most widely used illegal drug in the US. Escherichia coli was transformed with a plasmid containing a bacterial luxCDABEG bioluminescence gene cassette, activated by a cocaine-responsive signaling cascade. The engineered bioreporter is demonstrated to be a sensitive and specific first-generation detection system for cocaine, with detection thresholds of 17 ± 8 μg/L and 130 ± 50 μg/L in a buffer solution and in urine, respectively. Further improvement of the sensor’s performance was achieved by altering the nucleotide sequence of the PBen gene promoter, the construct’s sensing element, using accelerated site-directed evolution. The applicability of ready-to-use paper strips with immobilized bioreporter cells was demonstrated for cocaine detection in aqueous solutions.

Detection of tea leaf blight in UAV remote sensing images by integrating super-resolution and detection networks.

Tea leaf blight (TLB) is a common disease of tea plants and is widely distributed in tea gardens. Although the use of unmanned aerial vehicle (UAV) remote sensing can help to achieve a wider scale for TLB detection, the blurring of UAV images, overlapping of tea leaves, and small size of TLB spots pose significant challenges to the task of detection. This study proposes a method of detecting TLB in UAV remote sensing images by integrating super-resolution (SR) and detection networks. We use an SR network called SERB-Swin2sr to reconstruct the detailed features of UAV images and solve the problem of detail loss caused by the blurring in UAV images. In SERB-Swin2sr, a squeeze-and-excitation ResNet block (SERB) is introduced to enhance the models' ability to extract the target details in the images, and the convolution stem replaces the convolution block in order to increase the convergence rate and stability of the network. A detection network called SDDA-YOLO is applied to achieve precise detection of TLB in UAV remote sensing images. In SDDA-YOLO, a shuffle dual-dimensional attention (SDDA) module is introduced to enhance the feature fusion capability of the network, and an Xsmall-scale detection layer is used to enhance the detection ability of small lesions. Experimental results show that the proposed method is superior to current detection methods. Compared with a baseline YOLOv8 model, the precision, mAP@0.5, and mAP@0.5:0.95 of the proposed method are improved by 4.2%, 1.6%, and 1.8%, and the size of our model is only 4.6MB.

A Leakage Safety Discrimination Model and Method for Saline Aquifer CCS Based on Pressure Dynamics

The saline aquifer CCS is a crucial site for carbon storage. Safety monitoring is a key technology for saline aquifer CCS. Current CO2 leakage detection methods include microseismic, electromagnetic, and well-logging techniques. However, these methods face challenges, such as difficulties in determining CO2 migration fronts and predicting potential leakage events; as a result, the formulation of test timing and methods for these safety monitoring techniques are somewhat arbitrary. This study establishes a gas–water two-phase seepage model and solves it using a semi-analytical method to obtain the injection pressure and the derivative curve characteristics of the injection well. The pressure derivative curve can reflect the physical properties of the reservoir through which CO2 flows underground, and it can also be used to determine whether CO2 leakage has occurred, as well as the timing and amount of leakage, based on boundary responses. This study conducted sensitivity analyses on eight parameters to determine the impact of each parameter on the bottom-hole pressure and its derivatives, thereby obtaining the influence of its parameters on different flow stages. The research indicates that, when a steady-state flow characteristic appears at the outer boundary, CO2 leakage will occur. Additionally, the leakage location can be determined by calculating the distance from the injection well. This can guide the placement and measurement of safety monitoring methods for saline aquifer CCS. The method proposed in this paper can effectively monitor the timing, location, and amount of leakage, providing a technical safeguard for promoting CCS technology.

Pest Detection Based on Lightweight Locality-Aware Faster R-CNN

Accurate and timely monitoring of pests is an effective way to minimize the negative effects of pests in agriculture. Since deep learning-based methods have achieved good performance in object detection, they have been successfully applied for pest detection and monitoring. However, the current pest detection methods fail to balance the relationship between computational cost and model accuracy. Therefore, this paper proposes a lightweight, locality-aware faster R-CNN (LLA-RCNN) method for effective pest detection and real-time monitoring. The proposed model uses MobileNetV3 to replace the original backbone, reduce the computational complexity, and compress the size of the model to speed up pest detection. The coordinate attention (CA) blocks are utilized to enhance the locality information for highlighting the objects under complex backgrounds. Furthermore, the generalized intersection over union (GIoU) loss function and region of interest align (RoI Align) technology are used to improve pest detection accuracy. The experimental results on different types of datasets validate that the proposed model not only significantly reduces the number of parameters and floating-point operations (FLOPs), but also achieves better performance than some popular pest detection methods. This demonstrates strong generalization capabilities and provides a feasible method for pest detection on resource-constrained devices.

Electrochemical detection of human papillomavirus DNA based on an ultrasensitive electrochemical sensing platform using N-graphene paper

Human papillomavirus (HPV) is the primary cause of cervical cancer, a major global health concern affecting women worldwide. Early detection of high-risk HPV genotypes is crucial for timely intervention and reducing disease burden. However, current detection methods often lack sensitivity, speed, or cost-effectiveness, especially in resource-limited settings. This work addresses this critical need by developing an ultrasensitive electrochemical biosensing platform based on nitrogen-doped graphene paper electrodes for quantifying high-risk HPV16 DNA sequences. The N-graphene nanohybrid, fabricated via simple solvothermal treatment of biomass-derived graphene oxide, demonstrated significantly enhanced electrical and electrocatalytic properties compared to undoped graphene. DNA detection was achieved through characteristic oxidation signals of guanine bases, which displayed a wide linear dynamic range (0.5–100 μg/mL), low detection limit (75 pg/mL), and excellent reproducibility (relative standard deviation <3.5 %). Detailed spectroscopic and electrochemical analyses revealed the key roles of pyridinic nitrogen defects in improving interfacial charge transfer kinetics and mitigating biofouling issues. This synergistic combination of high electrical conductivity and versatile surface functionality paves the way for equipment-free, point-of-care genosensor devices tailored for quantitative biomarker screening in resource-constrained environments.

GNSS spoofing detection based on frequency domain processing

Global Navigation Satellite System (GNSS) spoofing detection has attracted much attention in recent years to ensure security. However, most of the existing methods use the time domain feature and can be easily affected by interference, which limits the practical application. To solve this problem, the frequency domain processing approach is investigated and a novel spoofing detection method is proposed. The basic idea is to exploit the spoofing-induced feature by transforming the correlator output into the frequency domain. The proposed method has a simple implementation logic and requires only minor software upgrades to existing GNSS receivers. Simulation and experimental results are presented to verify the effectiveness of the proposed method. The results demonstrate that it exhibits superior detection performance and shows robustness to interference. Consequently, this method has great application value and can serve as a valuable complement to current detection methods.