High Error Rate Research Articles

An FTwNB Shield: A Credit Risk Assessment Model for Data Uncertainty and Privacy Protection

Credit risk assessment is an important process in bank financial risk management. Traditional machine-learning methods cannot solve the problem of data islands and the high error rate of two-way decisions, which is not conducive to banks’ accurate credit risk assessment of users. To this end, this paper establishes a federated three-way decision incremental naive Bayes bank user credit risk assessment model (FTwNB) that supports asymmetric encryption, uses federated learning to break down data barriers between banks, and uses asymmetric encryption to protect data security for federated processes. At the same time, the model combines the three-way decision methods to realize the three-way classification of user credit (good, bad and delayed judgment), so as to avoid the loss of bank interests caused by the forced division of uncertain users. In addition, the model also incorporates incremental learning steps to eliminate training samples with poor data quality to further improve the model performance. This paper takes German Credit data and Default of Credit Card Clients data as examples to conduct simulation experiments. The result shows that the performance of the FTwNB model has been greatly improved, which verifies that it has good credit risk assessment capabilities.

Detecting lies in investigative interviews through the analysis of response latencies and error rates to unexpected questions

In this study, we propose an approach to detect deception during investigative interviews by integrating response latency and error analysis with the unexpected question technique. Sixty participants were assigned to an honest (n = 30) or deceptive group (n = 30). The deceptive group was instructed to memorize the false biographical details of a fictitious identity. Throughout the interviews, participants were presented with a randomized sequence of control, expected, and unexpected open-ended questions about identity. Responses were audio recorded for detailed examination. Our findings indicate that deceptive participants showed markedly longer latencies and higher error rates when answering expected (requiring deception) and unexpected questions (for which premeditated deception was not possible). Longer response latencies were also observed in participants attempting deception when answering control questions (which necessitated truthful answers). Moreover, a within-subject analysis highlighted that responding to unexpected questions significantly impaired individuals’ performance compared to answering control and expected questions. Leveraging machine-learning algorithms, our approach attained a classification accuracy of 98% in distinguishing deceptive and honest participants. Additionally, a classification analysis on single response levels was conducted. Our findings underscore the effectiveness of merging response latency metrics and error rates with unexpected questioning as a robust method for identity deception detection in investigative interviews. We also discuss significant implications for enhancing interview strategies.

The Consulting Research of ABC Warehouse

After discovering the existing problem in the HK distribution center of ABC, our research team got in touch with the staff of the warehouse and then completed the diagnosis of the problem as well as the consideration of optimization actions through the interview, combined with literature analysis. Our team interviewed 4 employees and summarized the current workflow and existing workarounds in the warehouse. In the current working environment, employees are under high work pressure, a high error rate, and the workflow does not match the work system. We put forward the system optimization suggestions for integrating Personal Digital Assistant (PDA), Warehouse Management System (WMS), and Ultra Wide Band (UWB) positioning technology and make corresponding adjustments in both equipment and management, hoping to get the expected outcomes of the intervention.

Identifying excessive length of antibiotic treatment duration for hospital-acquired infections: a semi-automated approach to support antimicrobial stewardship

BackgroundAvoiding excessive antibiotic treatment duration is a fundamental goal in antimicrobial stewardship. Manual collection of data is a time-consuming process, but a semi-automated approach for data extraction has been shown feasible for community-acquired infections (CAI). Extraction of data however may be more challenging in hospital-acquired infections (HAI). The aim of this study is to explore whether semi-automated data extraction of treatment duration is also feasible and accurate for HAI.MethodsData from a university-affiliated hospital over the period 1-6-2020 until 1-6-2022 was used for this study. From the Electronic Health Record, raw data on prescriptions, registered indications and admissions was extracted and processed to define treatment courses. In addition, clinical notes including prescription instructions were obtained for the purpose of validation. The derived treatment course was compared to the registered indication and the actual length of treatment (LOT) in the clinical notes in a random sample of 5.7% of treatment courses, to assess the accuracy of the data for both CAI and HAI.ResultsIncluded were 10.564 treatment courses of which 73.1% were CAI and 26.8% HAI. The registered indication matched the diagnosis as recorded in the clinical notes in 79% of treatment courses (79.2% CAI, 78.5% HAI). Higher error rates were seen in urinary tract infections (UTIs) (29.0%) and respiratory tract infections (RTIs) (20.5%) compared to intra-abdominal infections (7.4%), or skin or soft tissue infections (11.1%), mainly due to incorrect specification of the type of UTI or RTI. The LOT was accurately extracted in 98.5% of courses (CAI 98.2%, HAI 99.3%) when compared to prescriptions in the EHR. In 21% of cases however the LOT did not match with the clinical notes, mainly if patients received treatment from other health care providers preceding or following the present course.ConclusionSemi-automatic data extraction can yield reliable information about the indication and LOT in treatment courses of hospitalized patients, for both HAI and CAI. This can provide stewardship programs with a surveillance tool for all in-hospital treated infections, which can be used to achieve stewardship goals.

Lane detection using hough transformation and Yolov8

Autonomous vehicles necessitate the integration of advanced technologies such as computer vision and deep learning to comprehend and navigate their surroundings. A crucial yet challenging component of this integration is the accurate detection of lanes, which can be influenced by a multitude of varying lane characteristics and conditions. This research undertakes a comparative analysis of lane detection methodologies, explicitly focusing on traditional image processing techniques and Convolutional Neural Networks (CNNs). The evaluation utilized a sample of 500 images from the CULane dataset, which encompasses a diverse range of traffic scenarios. Initially, a method incorporating Gaussian blurring, Canny edge detection, and Hough line transformation was examined. Despite its efficiency, operating at 30 frames per second, this approach exhibited a high error rate (average Mean Squared Error (MSE) of 0.537), which is attributable to the loss of critical image details during the preprocessing stage. Subsequently, the performance of a fine-tuned YOLOv8 model, trained on a reformatted version of the CULane dataset was assessed. The combination of object detection and subsequent Hough transformation yielded high accuracy, demonstrating the model’s ability to learn and identify relevant lane features. The deep CNNs demonstrated superior performance over classical image processing techniques in terms of lane detection accuracy, thereby underscoring their potential applicability within the realm of autonomous vehicle technology

The application of artificial intelligence in Unmanned Underwater Vehicle communication systems

Artificial Intelligence (AI) has the potential to significantly enhance decision-making techniques in Unmanned Underwater Vehicle (UUV) communication systems by utilizing real-time sensor feedback and environmental data. To address the challenges posed by high data volume and error rates, we propose an advanced methodology known as the Variational Deep Network (VDN). This method integrates Variational Modal Decomposition (VMD) algorithms with the Deep Convolutional Neural Network (DCNN) to create a realizable AI decision-making system. In conjunction with the DCNN and VMD algorithms, our research focuses on accurately identifying received signals under data constraints. Considering the spectral efficiency and robustness, the Filter-Bank Multicarrier (FBMC) technology have been employed in this paper. Our results demonstrate that the VDN significantly reduces the volume of data needed for neural network training and effectively recognizes FBMC signals. These insights confirm the potential of VDN decision-making systems to drive advancements in UUV communication technologies.

Real-time detection of surface cracking defects for large-sized stamped parts

This study presents a framework for the real-time detection of surface cracking in large-sized stamped metal parts. The framework aims to address the challenges of low detection efficiency and high error rates associated with manual cracking detection. Within this framework, a novel network, SNF-YOLOv8, is proposed to efficiently detect cracking while ensuring that the detection speed matches the production speed. The network incorporates a convolutional spatial-to-depth module to enhance the detection of small-sized cracking and mitigate surface interference during inspections. Furthermore, a visual self-attention mechanism is introduced to improve feature extraction. A combination of standard convolutional and depth-wise separable convolutional layers in the neck network enhances speed without compromising accuracy. Experimental validation conducted using a dataset from actual production lines, in collaboration with a multi-national corporation, demonstrates that SNF-YOLOv8 achieves an average precision of 85.2% at a detection speed of 164 frames per second. The framework achieves an accuracy rate of 98.8% in detecting large-sized cracking and 96.4% in detecting small-sized cracking, meeting the requirements for high-precision and real-time detection applications.

Performance Comparison of Five Methods Available in ImageJ for Bird Counting and Detection from Video Datasets

Bird monitoring is an important approach to studying the diversity and abundance of birds, especially during migration, as it can provide core data for bird conservation purposes. The previous methods for bird number estimation are largely based on manual counting, which suffers from low throughput and a high error rate. In this study, we aimed to provide an alternative bird-counting method from video datasets by using five available ImageJ methods: Particle Analyzer, Find Maxima, Watershed segmentation, TrackMate, and trainable WEKA segmentation. The numbers of birds and their XY coordinates were extracted from videos to conduct a side-by-side comparison with the manual counting results, and the three important criteria of the sensitivity, precision, and F1 score were calculated for the performance evaluation. From the tests, which we conducted for four different cases with different bird numbers or flying patterns, TrackMate had the best overall performance for counting birds and pinpointing their locations, followed by Particle Analyzer, Find Maxima, WEKA, and lastly, Watershed, which showed low precision in most of the cases. In summary, five ImageJ-based counting methods were compared in this study, and we validated that TrackMate obtains the best performance for bird counting and detection.

Optimizing the Use of Dose Error Reduction Software on Intravenous Infusion Pumps.

Intravenous infusions have the potential to cause significant harm in patients and are associated with a high rate of adverse drug events and medication errors. Infusion pumps with dose error reduction software (DERS) can be used to reduce errors by establishing safe infusion parameters. In 2019, a quality improvement project was initiated with the aim to increase DERS compliance from 46% to 75% at our specialty institution by October 1, 2022. An interdisciplinary group was tasked with improving compliance with DERS by identifying key drivers, including informed staff, engaged staff, and an accurate smart pump library. We used the Model for Improvement framework to guide this improvement project, and Plan-Do-Study-Act (PDSA) cycles were used to plan for interventions. PDSA cycles included drug library updates, education, and unit-level compliance reporting. Weekly average DERS compliance was monitored as the outcome measure, and weekly pump alerts per 100 infusions were monitored as a balancing measure; statistical process control charts were used to monitor measures from 2018 to 2022. Over the course of 25 months, 8 PDSA cycles resulted in 5 centerline improvements from a baseline mean of 46% to a final mean of 78%. Pump alerts per 100 infusions decreased from 15.9 to 6.4 with the first PDSA cycle and then continued to decrease to 3.9 with subsequent interventions. Although features like DERS can help ensure safe medication administration, continuous improvement efforts to increase DERS compliance without increasing alert burden are needed to ensure that benefits of this technology are optimized.

The Association Between Task Complexity and Cortical Language Mapping Accuracy.

Direct cortical stimulation (DCS) mapping enables the identification of functional language regions within and around gliomas before tumor resection. Intraoperative mapping is required because glioma-infiltrated cortex engages in synchronous activity during task performance in a manner similar to normal-appearing cortex but has decreased ability to encode information for complex tasks. It is unknown whether task complexity influenced DCS mapping results. We aim to understand correlations between audiovisual picture naming (PN) task complexity and DCS error rate. We also asked what functional and oncological factors might be associated with higher rates of erroneous responses. We retrospectively reviewed intraoperative PN and word reading (WR) task performance during awake DCS language mapping for resection of dominant hemisphere World Health Organization grade 2 to 4 gliomas. The complexity of word tested in PN/WR tasks, patient characteristics, and tumor characteristics were compared between correct and incorrect trials. Between 2017 and 2021, 74 patients met inclusion criteria. At median 18.6 months of follow-up, 73.0% were alive and 52.7% remained recurrence-free. A total of 2643 PN and 978 WR trials were analyzed. A greater number of syllables in PN was associated with a higher DCS error rate (P = .001). Multivariate logistic regression found that each additional syllable in PN tasks independently increased odds of error by 2.40 (P < .001). Older age was also an independent correlate of higher error rate (P < .043). World Health Organization grade did not correlate with error rate (P = .866). More severe language impairment before surgery correlated with worse performance on more complex intraoperative tasks (P < .001). A higher error rate on PN testing did not correlate with lower extent of glioma resection (P = .949). Word complexity, quantified by the number of syllables, is associated with higher error rates for intraoperative PN tasks but does not affect extent of resection.

Algebraic side‐channel attacks on Trivium stream cipher

AbstractAlgebraic Side‐Channel Attacks (ASCAs), first proposed by Renauld and Standaert in 2009, are a potent cryptanalysis method against block ciphers. In this paper, the authors initially utilize ASCAs to analyze the security of the Trivium stream cipher, given its concise algebraic structure. Considering its efficiency in both hardware and software implementations, the authors deploy ASCAs to target Trivium implemented both in application specific integrated circuit (ASIC) under the Hamming Distance Leakage Model (HDLM) (noted as CASE 1) and in microcontrollers of various buses (i.e. some common 8‐bit, 16‐bit, and 32‐bit architectures, noted as CASE 2, CASE 3, and CASE 4, respectively) under the Hamming Weight Leakage Model (HWLM). Here, the authors’ attacks are conducted on power‐simulated targets and not on real devices. For a single power consumption trace without measurement errors, this paper presents experimental results using MiniSat 2.0. Unfortunately, the authors were unable to break the ASIC implementation of Trivium under HDLM (CASE 1) with a time complexity of 2109 s or so, which is worse than the exhaustive key attack. For CASEs 2 to 4, the authors can find the complete 288‐bit state of Trivium within a reasonable timeframe. Specially, the success rate can reach 100% with an average solving time of less than 1 s when only measuring the leakages of the first eight consecutive rounds for CASE 2. Furthermore, the authors can still successfully recover the internal state even when obtaining leakages of the first 41 rounds with a random loss rate. In fact, it can tolerate a 74% random loss rate for the first 223 rounds. With regard to the potential errors in the measurements, the authors mitigate them using Tolerant ASCA (TASCA). Similarly, CASE 1 cannot be compromised even in error‐free situations, while the authors can still successfully recover the internal state of CASEs 2 to 4 from a single power trace, even with a high error rate, including 100% incorrect measurements. Surprisingly, for CASEs 2 to 4, the authors can recover the internal state with a 100% success rate, regardless of the error rate. As a result, the security of Trivium will not be enhanced when transitioning from a smaller 8‐bit platform to a larger 32‐bit platform. In the end, the authors will consider some more abstract attack models. The results can provide us with additional insights into the security of Trivium from a different perspective.

Spatial sensitivity distribution assessment and Monte Carlo simulations for needle-based bioimpedance imaging during venipuncture using the finite element method.

Despite being among the most common medical procedures, needle insertions suffer from a high error rate. Impedance measurements using electrode-equipped needles offer promise for improved tissue targeting and reduced errors. Impedance visualization usually requires an extensive pre-measured impedance dataset for tissue differentiation and knowledge of the electric fields contributing to the resulting impedances. This work presents two finite element simulation approaches for both problems. The first approach describes the generation of a multitude of impedances with Monte Carlo simulations for both, homogeneous and inhomogeneous tissue to circumvent the need to rely on previously measured data. These datasets could be used for tissue discrimination. The second method describes the simulation of the spatial sensitivity distribution of an electrode layout. Two singularity analysis methods were employed to determine the bulk of the sensitivity within a finite volume, which in turn enables consistent 3D visualization. The modeled electrode layout consists of 12 electrodes radially placed around a hypodermic needle. Electrical excitation was simulated using two neighboring electrodes for current carriage and voltage pickup, which resulted in 12 distinct bipolar excitation states. Both, the impedance simulations and the respective singularity analysis methods were compared with each other. The results show that the statistical spread of impedances is highly dependent on the tissue type and its inhomogeneities. The bounded bulk of sensitivities of both methods are of similar extent and symmetry. Future models should incorporate more detailed tissue properties such as anisotropy or changing material properties due to tissue deformation to gain more accurate predictions.

Safeguarding cyberspace: Enhancing malicious website detection with PSO[sbnd]optimized XGBoost and firefly-based feature selection

In recent years, the exponential growth of internet usage worldwide has created a conducive environment for the expansion of malicious activities. Among these threats, malicious websites pose a significant risk to individual users and corporations. This paper presents a robust and efficient model for the detection of various types of malicious websites with high accuracy in the process. The proposed approach employs a two-step process. Firstly, a feature selection method based on the Firefly algorithm is utilized to identify the most relevant features. Subsequently, an optimized version of the XGBoost algorithm is applied to classify websites based on the selected features. Optimization of XGBoost's parameters is achieved through the Particle Swarm Optimization (PSO) algorithm to enhance its performance. To assess the efficacy of the introduced model, the model is evaluated against several benchmark classification algorithms using a dataset comprising over 36,000 websites. In binary classification, the introduced model surpasses other benchmark methods with a significant 98.42 % classification accuracy and an F1 score of 0.984. For multiclass problem classification, it consistently achieves over 98 % accuracy in each class. The test results highlight the proposed model's robust performance, characterized by exceptional classification accuracy and F1-measure rates. It demonstrates the capability to detect various types of malicious websites with high precision and minimal false error rates.

ENHANCING BRIX VALUE PREDICTION IN STRAWBERRIES USING MACHINE LEARNING: A FUSION OF PHYSIOCHEMICAL AND COLOR-BASED FEATURES FOR IMPROVED SWEETNESS ASSESSMENT

This study contributes to the ongoing wave of artificial intelligence integration by applying machine learning techniques to automate the assessment of strawberry quality. This research focuses on determining if the sweetness of strawberries can be predicted using a combination of physiochemical variables, their interaction parameters, and color-based features extracted from image data. This research used a 150-sample collection of strawberry images and physiochemical characteristics such as salinity, specific gravity, pH, and Brix. Normalized raw and derived feature variables and selected dataset transformations were done. We then split the dataset into mutual exclusivity training and test sets. Exponential Gaussian Process Regression (GPR) suited well due to low validation errors. This best model predicted Brix values for the remaining test samples. The Mean Absolute Percentage Error(MAPE) showed 98.783% forecast accuracy (Acc). We also examined the model's coefficient of determination (R2) values, which were 0.78 and 0.9739 for training and testing, respectively. The Mean Square Error (MSE) and Mean Absolute Error (MAE) obtained after training were 0.32994 and 0.0453, and testing was 0.35286 and 0.0663. Using input characteristics with high Acc and low error rates, deep learning models like Recurrent Neural Network (RNN) and its derivatives were constructed. Using physiochemical and visual data, machine learning and deep learning models successfully predict strawberry sweetness. This prediction accuracy shows the complex link between internal components and Brix readings, enabling high-quality strawberry production.

EA-GAT: Event aware graph attention network on cyber-physical systems

Anomaly detection with high accuracy, recall, and low error rate is critical for the safe and uninterrupted operation of cyber-physical systems. However, detecting anomalies in multimodal time series with different modalities obtained from cyber-physical systems is challenging. Although deep learning methods show very good results in anomaly detection, they fail to detect anomalies according to the requirements of cyber-physical systems. In the use of graph-based methods, data loss occurs during the conversion of time series into graphs. The fixed window size used to transform time series into graphs causes a loss of spatio-temporal correlations. In this study, we propose an Event Aware Graph Attention Network (EA-GAT), which can detect anomalies by event-based cyber-physical system analysis. EA-GAT detects and tracks the sensors in cyber-physical systems and the correlations between them. The system analyzes and models the relationship between the components during the marked periods as a graph. Anomalies in the system are found through the created graph models. Experiments show that the EA-GAT technique is more effective than other deep learning methods on SWaT, WADI, MSL datasets used in various studies. The event-based dynamic approach is significantly superior to the fixed-size sliding window technique, which uses the same learning structure. In addition, anomaly analysis is used to identify the attack target and the affected components. At the same time, with the slip subsequence module, the data is divided into subgroups and processed simultaneously.

NextDenovo: an efficient error correction and accurate assembly tool for noisy long reads

Long-read sequencing data, particularly those derived from the Oxford Nanopore sequencing platform, tend to exhibit high error rates. Here, we present NextDenovo, an efficient error correction and assembly tool for noisy long reads, which achieves a high level of accuracy in genome assembly. We apply NextDenovo to assemble 35 diverse human genomes from around the world using Nanopore long-read data. These genomes allow us to identify the landscape of segmental duplication and gene copy number variation in modern human populations. The use of NextDenovo should pave the way for population-scale long-read assembly using Nanopore long-read data.

Generating barcodes for nanopore sequencing data with PRO

DNA barcodes, short and unique DNA sequences, play a crucial role in sample identification when processing many samples simultaneously, which helps reduce experimental costs. Nevertheless, the low quality of long-read sequencing makes it difficult to identify barcodes accurately, which poses significant challenges for the design of barcodes for large numbers of samples in a single sequencing run. Here, we present a comprehensive study of the generation of barcodes and develop a tool, PRO, that can be used for selecting optimal barcode sets and demultiplexing. We formulate the barcode design problem as a combinatorial problem and prove that finding the optimal largest barcode set in a given DNA sequence space in which all sequences have the same length is theoretically NP-complete. For practical applications, we developed the novel method PRO by introducing the probability divergence between two DNA sequences to expand the capacity of barcode kits while ensuring demultiplexing accuracy. Specifically, the maximum size of the barcode kits designed by PRO is 2,292, which keeps the length of barcodes the same as that of the official ones used by Oxford Nanopore Technologies (ONT). We validated the performance of PRO on a simulated nanopore dataset with high error rates. The demultiplexing accuracy of PRO reached 98.29% for a barcode kit of size 2,922, 4.31% higher than that of Guppy, the official demultiplexing tool. When the size of the barcode kit generated by PRO is the same as the official size provided by ONT, both tools show superior and comparable demultiplexing accuracy.

A Stage-Wise Residual Attention Generation Adversarial Network for Mandibular Defect Repairing and Reconstruction.

Surgical reconstruction of mandibular defects is a clinical routine manner for the rehabilitation of patients with deformities. The mandible plays a crucial role in maintaining the facial contour and ensuring the speech and mastication functions. The repairing and reconstruction of mandible defects is a significant yet challenging task in oral-maxillofacial surgery. Currently, the mainly available methods are traditional digitalized design methods that suffer from substantial artificial operations, limited applicability and high reconstruction error rates. An automated, precise, and individualized method is imperative for maxillofacial surgeons. In this paper, we propose a Stage-wise Residual Attention Generative Adversarial Network (SRA-GAN) for mandibular defect reconstruction. Specifically, we design a stage-wise residual attention mechanism for generator to enhance the extraction capability of mandibular remote spatial information, making it adaptable to various defects. For the discriminator, we propose a multi-field perceptual network, consisting of two parallel discriminators with different perceptual fields, to reduce the cumulative reconstruction errors. Furthermore, we design a self-encoder perceptual loss function to ensure the correctness of mandibular anatomical structures. The experimental results on a novel custom-built mandibular defect dataset demonstrate that our method has a promising prospect in clinical application, achieving the best Dice Similarity Coefficient (DSC) of 94.238% and 95% Hausdorff Distance (HD95) of 4.787.

Exploration on Wireless Network-Based Monitoring and Early Warning Algorithms for Power Grounding Faults

Electricity is an indispensable resource in daily life. While it brings great convenience to people, it also brings some safety hazards, especially in the event of power failure, which may cause significant harm. Therefore, monitoring and warning of power faults are very important. Traditional power grounding fault monitoring and warning systems have problems such as untimely monitoring, inaccurate warning, and high error rates in fault location. In order to solve the above problems, this paper uses wireless networks to construct a power grounding fault monitoring and early warning system. The wireless network collected fault data based on the complexity of power grounding fault data, and used threshold monitoring method to analyze the collected data. The wireless network was used to construct a prediction model to monitor grounding faults. Through experiments, it can be found that the accuracy of the wireless network-based power fault monitoring system for predicting grounding faults was over 92.57%, and the average warning accuracy of 20 experiments was 93.856%. This paper studied a wireless network-based algorithm for monitoring and warning of power grounding faults, which can effectively improve the monitoring and warning capabilities of power systems, and reduce the risk of power equipment faults and the probability of power accidents.

De novo diploid genome assembly using long noisy reads

The high sequencing error rate has impeded the application of long noisy reads for diploid genome assembly. Most existing assemblers failed to generate high-quality phased assemblies using long noisy reads. Here, we present PECAT, a Phased Error Correction and Assembly Tool, for reconstructing diploid genomes from long noisy reads. We design a haplotype-aware error correction method that can retain heterozygote alleles while correcting sequencing errors. We combine a corrected read SNP caller and a raw read SNP caller to further improve the identification of inconsistent overlaps in the string graph. We use a grouping method to assign reads to different haplotype groups. PECAT efficiently assembles diploid genomes using Nanopore R9, PacBio CLR or Nanopore R10 reads only. PECAT generates more contiguous haplotype-specific contigs compared to other assemblers. Especially, PECAT achieves nearly haplotype-resolved assembly on B. taurus (Bison×Simmental) using Nanopore R9 reads and phase block NG50 with 59.4/58.0 Mb for HG002 using Nanopore R10 reads.