Test Accuracy Research Articles

Physiological Profile Assessment and Self-Measurement of Healthy Students through Remote Protocol during COVID-19 Lockdown

Background: The COVID-19 pandemic has led to reduced physical activity and increased sedentary behaviors, negatively impacting mental and physical health. Engaging in physical activity at home during quarantine became essential to counteracting these adverse effects. To develop appropriate physical activity programs, assessing individuals’ fitness levels and the impact of inactivity is crucial. This study aims to compare motor abilities—including flexibility, balance, reaction time, cardiovascular endurance, and lower and upper limb strength—assessed both in-person and remotely, to determine the accuracy and repeatability of self-administered tests. Methods: A total of 35 young subjects (age 24.2 ± 1.97 years, BMI 22.4 ± 2.61 kg/m2) participated in this study. Each participant underwent a battery of motor ability tests designed to assess various fitness components. The tests were administered twice for each subject: once in a laboratory setting and once remotely at home. The sequence of tests was randomly assigned to ensure unbiased results. Both the in-person and remote assessments were used to evaluate the accuracy and reliability of self-administered tests. Results: The comparison of test results between the laboratory and remote settings revealed percentage differences ranging from 5% to 10%. This variation is considered an acceptable margin of error, suggesting that the tests conducted remotely were relatively accurate when compared to those performed in a controlled laboratory environment. Conclusions: The findings indicate that remote fitness testing is a promising method for evaluating motor abilities. With an acceptable margin of error, remote assessments can be effectively used to personalize training programs based on individuals’ physiological characteristics. This approach may be particularly beneficial during times of limited access to fitness facilities, such as during quarantine, or for individuals seeking more flexible fitness evaluation methods.

Detecting malingered neurocognitive dysfunction: Comparative analysis of freestanding and embedded performance validity tests

Detecting malingered neurocognitive dysfunction is a major issue in forensic context, particularly in legal proceeding/insurance assessment after a traumatic brain injury (TBI), condition frequently associated with persistent cognitive impairments that may potentially be related to malingering. Consequently, research has devoted considerable efforts on developing tools to verify symptoms authenticity. This study compared two freestanding performance validity tests (PVTs) (Amsterdam Short-Term Memory Test—ASTM; Word Completion Memory Test—WCMT) and five embedded PVTs (Rey Complex Figure Test—RCFT—Copy and Recall trials; Reliable Digit Span—RDS; Rey Auditory Verbal Learning Test—RAVLT—Recognition and Total Learning trials) in a sample of 120 participants, including 15 patients with TBI (Mage = 44.40), 52 experimental simulators (Mage = 29.52) and 53 control (Mage = 29.77). Group performance was analyzed to assess tests’ discriminatory power, and Receiver Operating Characteristic (ROC) curves were used to examine tools’ sensitivity and specificity. Results indicated that experimental simulators performance on the ASTM, WCMT, and RAVLT differed significantly from TBI patients and controls. The RDS and RCFT did not discriminate experimental simulators from TBI group. ROC curves analysis reveals that the most accurate tests in this battery for detecting malingering were the ASTM and the RAVLT-Total Learning score. This study offers implications for identifying malingering in medico-legal settings, underscoring the importance of incorporating PVTs into clinical practice.

Sensor-Embedded Tissue Phantom for Magnetic Resonance Elastography Mechanical Failure Testing

Abstract Magnetic Resonance Elastography (MRE) is an imaging technique capable of quantifying the stiffness of in vivo tissue by applying and imaging shear waves produced by an MRE actuator. Poor image acquisition may result from the MRE procedure if there is insufficient contact between the MRE actuator and the patient. An experimental test setup outside of the clinic will aid in reducing the number of failed acquisitions by enabling the development of advanced actuators and actuator systems. This work presents the development and testing of a sensor-embedded tissue phantom setup paired with a support vector machine (SVM) classifier to automate the MRE actuator testing process. MRE actuation of soft tissue is simulated by utilizing a voice coil positioning stage that interfaces with a phantom. To capture the resulting vibrations, accelerometers are embedded inside the phantom. Subsequent characterization experiments verify the functionality of the developed phantoms to capture wave propagation. A secondary investigation was performed by utilizing the developed setup to collect acceleration measurements at varying contact distances. We provide an overview of feature analysis and selection to develop SVM models for contact detection. Multiple SVM models are reported, and the best-performing model displayed almost perfect validation (94.53%) and test (90.91%) accuracy. The pairing of sensor-embedded phantom with an SVM for detection demonstrates potential improvements to the MRE actuator developmental process by automatically assessing contact-related issues prior to clinical testing.

Diagnostic and Predictive Immunocytochemistry in Lung Cancer

Background Immunocytochemistry (ICC) is suitable for use on a range of cytology preparations, such as cell blocks, air-dried slides, ethanol-fixed slides, direct smears, cytospins, and liquid-based cytology (LBC) samples. However, it must be standardized against the gold standard of formalin-fixed paraffin-embedded tissues with adequate number of positive and negative controls. The role of ICC in lung cancer is crucial, as most lung cancer specimens are cytology samples. Accurate diagnosis and testing of certain biomarkers rely heavily on both diagnostic and predictive ICC. Summary Key ICC markers important in lung cancer include, but are not limited to, diagnostic ICCs such as TTF-1, p40, Napsin A, and p63, as well as predictive ICCs like ALK, ROS-1, PD-L1, and NTRK. Key Messages With proper validation, immunocytochemistry for lung cancer can be effectively performed on direct smears, cytospins, and other specimens, even when resources for preparing cell blocks are unavailable. This is particularly true for diagnostic antibodies, but it is important to exercise caution with predictive immunocytochemistry. Nonetheless, a low threshold for molecular testing should be maintained. PD-L1 ICC can be challenging and should ideally be performed on formalin-fixed cell blocks or biopsies when available.

Analyzing the impact of transfer learning on explanation accuracy in deep learning-based ECG recognition systems

Abstract Electrocardiogram (ECG) recognition systems now play a leading role in the early detection of cardiovascular diseases. However, the explanation of judgments made by deep learning models in these systems is prominent for clinical acceptance. This article reveals the effect of transfer learning in ECG recognition systems on decision precision. This article investigated the role of transfer learning in ECG image classification using a customized convolutional neural network (CNN) with and without a VGG16 architecture. The customized CNN model with the VGG16 achieved a good test accuracy of 98.40%. Gradient-weighted class activation mapping (Grad-CAM), for this model, gave the wrong information because it focused on parts of the ECG that were not important for making decisions instead of features necessary for clinical diagnosis, like the P wave, QRS complex, and T wave. A proposed model that only used customized CNN layers and did not use transfer learning performed 99.08% on tests gave correct Grad-CAM explanations and correctly identified the influencing areas of decision-making in the ECG image. Because of these results, it seems that transfer learning might provide good performance metrics, but it might also make things harder to understand, which could make it harder for deep learning models that use ECG recognition to be reliable for diagnosis. This article concludes with a call for careful consideration when using transfer learning in the medical field, as model explanations resulting from such learning may not be appropriate when it comes to domain-specific interpretations.

Immunity to Non-Dengue Flaviviruses Impacts Dengue Virus Immunoglobulin G Enzyme-Linked Immunosorbent Assay Specificity in Cambodia

Abstract Background Seroprevalence studies are the standard for disease surveillance, and serology determined eligibility for the first dengue vaccine. Expanding flavivirus co-circulation and vaccination complicate testing. We evaluate the accuracy of a common dengue virus serological assay, examine immunity to non-dengue flaviviruses as a contributor to decreased performance, and assess whether alternative cut points may improve assay performance. Methods Children (n = 770) aged 2–9 years in Kampong Speu, Cambodia were enrolled in a prospective longitudinal study, and PanBio indirect dengue virus immunoglobulin G (IgG) enzyme-linked immunosorbent assay (ELISA) was performed. Plaque reduction neutralization tests (PRNTs) using dengue viruses were performed on a subset to assess the accuracy of the IgG ELISA, and PRNTs with Zika, Japanese encephalitis, and West Nile viruses evaluated immunity to non-dengue flaviviruses. Receiver operating curve analysis identified an alternative cut point to improve IgG ELISA accuracy. Results The dengue IgG ELISA had a lower specificity than previously reported (58% vs 93%–100%). Of those with false-positive IgG results, 46% had detectable neutralizing antibodies against other flaviviruses including 14% against West Nile virus. A higher IgG cut point improved the test accuracy in this population. Conclusions Physicians and public health authorities should be alert for West Nile in Cambodia. Immunity to non-dengue flaviviruses can impact dengue surveillance. Clinical Trials Registration NCT03534245.

Error analysis model-driven workflow for self-calibration stitching testing of X-ray flat surfaces

Self-calibration stitching test is widely utilized for testing X-ray flat surfaces. Various factors, including environment disturbances, motion errors, misalignments, overlapping ratios, and sub-aperture sizes can affect the accuracy. Previous studies have shown that motion errors can introduce slope errors in the test surface, and higher overlapping ratios can increase the test error. However, these error analyses have not been comprehensive or quantitatively sufficient for practical applications. For X-ray flat surfaces with specific accuracy requirements, the necessary control levels for error sources and the optimal test parameters remain unclear. To this end, an error analysis model-driven workflow for self-calibration stitching testing of X-ray flat surfaces is proposed. This model allows for the systematic evaluation of how different error sources and test parameters affect test accuracy. Further, it can guide the determination of the control levels of error sources and test parameters, moving beyond the reliance on empirical experience as seen in previous studies. A flat X-ray mirror with clear aperture of 250 mm × 30 mm and 0.2 nm (50 nrad) RMS accuracy requirement was utilized to demonstrate the procedures of the workflow. Verification experiments demonstrated that this workflow can serve as a standard model for evaluating test accuracy and guiding test procedures to achieve the desired accuracy in self-calibration stitching of X-ray flat mirrors.

Diagnostic performance of pupil perimetry in detecting hemianopia under standard and virtual reality viewing conditions.

To determine the diagnostic performance and reliability of two pupil perimetry (PP) methods in homonymous hemianopia. This cross-sectional monocenter cohort study performed gaze-contingent flicker PP (gcFPP) and a virtual reality version of gcFPP (VRgcFPP) twice on separate occasions in all patients suffering from homonymous hemianopia due to neurological impairment. The main outcomes were (1) test accuracy and (2) test-retest reliability: (1) was measured through area under the receiver operating characteristics curve (AUC) calculation of (VR)gcFPP results with comparators being SAP and healthy controls, respectively; (2) was evaluated by comparing tests 1 and 2 of both methods within patients. Both gcFPP and VRgcFPP were performed in 15 patients (12 males, MAge = 57, SDAge = 15) and 17 controls (6 males, MAge = 53, SDAge = 12). Mean test accuracy was good in separating damaged from intact visual field regions (gcFPP: Mauc = 0.83, SDauc = 0.09; VRgcFPP: Mauc = 0.69,SDauc = 0.13) and in separating patients from controls (gcFPP: Mauc = 0.92,SDauc = 0.13; VRgcFPP: Mauc = 0.96,SDauc = 0.15). A high test-retest reliability was found for the proportion intact versus damaged visual field (gcFPP: r = 0.95, P < .001, VRgcFPP: r = 1.00, P < .001). Overall, these results can be summarized as follows: (1) the comparison of pupil response amplitudes between intact versus damaged regions per patient indicate that gcFPP allows for cleaner imaging of intact versus damaged visual field regions than VRgcFPP, (2) the comparisons of average differences in intact versus damaged amplitudes between patients and controls demonstrate high diagnostic performance of both gcFPP and VRgcFPP, and (3) the test-retest reliabilities confirm that both gcFPP and VRgcFPP reliably and consistently measure defects in homonymous hemianopia. KEY MESSAGES: What is known Standard automated perimetry is the current gold standard for visual field examination, but not always suited for the evaluation of the VF in neurologically impaired patients. Pupil perimetry consists of the measurement of pupillary responses to light stimuli as a measure of visual sensitivity. What is new This study reports the highest diagnostic accuracy of pupil perimetry so far in patients with homonymous hemianopia. Gaze-contingent flicker pupil perimetry reliably and consistently measures defects in homonymous hemianopia under standard and virtual reality viewing conditions.

Machine learning algorithm for predicting seizure control after temporal lobe resection using peri-ictal electroencephalography

Brain resection is curative for a subset of patients with drug resistant epilepsy but up to half will fail to achieve sustained seizure freedom in the long term. There is a critical need for accurate prediction tools to identify patients likely to have recurrent postoperative seizures. Results from preclinical models and intracranial EEG in humans suggest that the window of time immediately before and after a seizure (“peri-ictal”) represents a unique brain state with implications for clinical outcome prediction. Using a dataset of 294 patients who underwent temporal lobe resection for seizures, we show that machine learning classifiers can make accurate predictions of postoperative seizure outcome using 5 min of peri-ictal scalp EEG data that is part of universal presurgical evaluation (AUC 0.98, out-of-group testing accuracy > 90%). This is the first approach to seizure outcome prediction that employs a routine non-invasive preoperative study (scalp EEG) with accuracy range likely to translate into a clinical tool. Decision curve analysis (DCA) shows that compared to the prevalent clinical-variable based nomogram, use of the EEG-augmented approach could decrease the rate of unsuccessful brain resections by 20%.

Construction and validation of a risk prediction model for extrauterine growth restriction in preterm infants born at gestational age less than 34 weeks

ObjectiveThis study aimed to develop and validate a model for predicting extrauterine growth restriction (EUGR) in preterm infants born ≤34 weeks gestation.MethodsPreterm infants from Guangxi Maternal and Child Health Hospital (2019–2021) were randomly divided into training (80%) and testing (20%) sets. Collinear clinical variables were excluded using Pearson correlation coefficients. Predictive factors were identified using Lasso regression. Random forest (RF), support vector machine (SVM), and logistic regression (LR) models were then built and evaluated using the confusion matrix, area under the curve (AUC), and the F1 score. Additionally, calibration curves and decision curve analysis (DCA) were plotted to assess the performance and practical utility of the models.ResultsThe study included 387 infants, with no significant baseline differences between training (n = 310) and testing (n = 77) sets. LR identified gestational age, birth weight, premature rupture of membranes, patent ductus arteriosus, cholestasis, and neonatal sepsis as key EUGR predictors. The RF model (19 variables) demonstrated an accuracy of greater than 90% during training, and superior AUC (0.62), F1 score (0.80), and accuracy (0.72) in testing compared to other models.ConclusionsGestational age, birth weight, premature rupture of membranes, patent ductus arteriosus, cholestasis, and neonatal sepsis are significant EUGR predictors in preterm infants ≤34 weeks. The model shows promise for early EUGR prediction in clinical practice, potentially enhancing screening efficiency and accuracy, thus saving medical resources.

Novel loss-of-function variants in WDR26 cause Skraban-Deardorff syndrome in two Chinese patients

IntroductionMutations in the protein WD repeat structural domain 26 (WDR26, MIM 617424) have been identified as the cause of autosomal dominant Skraban-Deardorff syndrome, a rare genetic disorder characterized by intellectual disability (ID), developmental delay (DD), hypotonia, epilepsy, infant feeding difficulties, gait abnormalities and distinctive facial features. The objective of this study is to investigate the genetic factors that may contribute to the development of Skraban-Deardorff syndrome in affected individuals.MethodsIn this study, we used whole-exome sequencing (WES) to analyze pathogenic and likely pathogenic variants in two unrelated Chinese patients with DD and ID. We confirmed the origin of the variants by conducting Sanger sequencing and classified them according to ACMG/AMP guidelines.ResultsHere, two novel de novo variants (c.1797delC(p.His599fs*11) and c.1414C&amp;gt;T(p.Gln472*)) in the WDR26 gene have been identified in two Chinese patients with Skraban-Deardorff syndrome. These patients exhibit a range of symptoms, including varying degrees of ID, DD, speech delay, an abnormal wide-foot and/or stiff-legged gait, facial dysmorphism, behavioural abnormalities, with or without seizures.ConclusionsIn this study, We report two unrelated Chinese patients with Skraban-Deardorff syndrome caused by novel de novo pathogenic variants of the WDR26 gene. These patients showed a clinical phenotype similar to that of patients with the WDR26 variant. Compared to reported cases with WDR26 pathogenic variants, patient 2 presented a novel complication of severe behavioural problems, including hyperactivity, social anxiety, self-mutilation, impulsivity and violent behaviour. This research broadens the range of genetic and clinical features of Skraban-Deardorff syndrome. In addition, the symptoms may become more pronounced as the patient ages. Furthermore, our report highlights the clinical diversity of Skraban-Deardorff syndrome. The findings may assist healthcare professionals in providing more accurate genetic testing and counselling to affected families and improving the overall management of the condition.

Landslide Analysis with Incomplete Data: A Framework for Critical Parameter Estimation

Landslides are one of the most common geohazards, posing significant risks to infrastructure, recreation, and human life. Slope stability analyses rely on detailed data, accurate materials testing, and careful model parameter selection. These factors are not always readily available, and estimations must be made, introducing uncertainty and error to the final slope stability analysis results. The most critical slope stability parameters that are often missing or incompletely constrained include slope topography, depth to water table, depth to failure plane, and material property parameters. Though estimation of these values is common practice, there is limited guidance or best practice instruction for this important step in the analysis. Guidance is provided for the estimation of: original and/or post-failure slope topography via traditional methods as well as the use of open-source digital elevation models, water table depth across variable hydrologic settings, and the iterative estimation of depth to failure plane and slope material properties. Workflows are proposed for the systematic estimation of critical parameters based primarily on slide type and scale. The efficacy of the proposed estimation techniques, uncertainty quantification, and final parameter estimation protocol for data-sparse landslide analysis is demonstrated via application at a landslide in Colorado, USA.

Comparative evaluation of plasma biomarkers of Schistosoma haematobium infection in endemic populations from Burkina Faso.

Infection with Schistosoma haematobium causes urogenital disease associated with organ disfunction, bleeding, pain, and higher susceptibility to infections and cancer. Timely and accurate diagnosis is crucial for prompt and appropriate treatment as well as surveillance efforts, and the use of plasma biomarkers offers important advantages over parasitological examination of urine, including increased sensitivity and the possibility to use the same specimen for multiple investigations. The present study aims to evaluate the diagnostic performance of different plasma biomarkers in endemic populations from Burkina Faso, West Africa. Schistosoma spp. Circulating Anodic Antigen (CAA), cell free S. haematobium DNA (cfDNA), class M and G antibodies against S. haematobium Soluble Worm Antigen Preparation (SWAP) and Soluble Egg Antigen (SEA) were measured in 406 plasma samples. Results of each biomarker test were compared to those of CAA, a Composite Reference Standard (CRS) and Latent Class Analysis (LCA). An identical proportion of positive samples (29%) was observed as a result of CAA and cfDNA testing, with a substantial agreement (84%, Cohen k = 0.62) between the results of the two tests, and a comparable agreement with the results of CRS and LCA. A higher positivity was observed, as expected, as a result of specific antibody testing (47%-72%), with IgG showing a higher agreement than IgM with the three references. Also, higher IgG levels were observed in current vs past infection, and ROC analysis identified optimal cutoff values for improved testing accuracy. This study provides compelling evidence that can inform the choice of the most appropriate diagnostic plasma biomarker for urogenital schistosomiasis in endemic areas, depending on the purpose, context, and available resources for testing. Either CAA or cfDNA testing can be used for the diagnosis of patients and for epidemiological investigations, even in absence of urine filtration microscopy, whereas anti-SWAP or anti-SEA IgG can be employed for surveillance and integrated monitoring of control interventions against poverty-associated diseases.

Requirements for partial turbulence simulations using nondimensional turbulence energy contributions

Quantifying turbulence effects is crucial for understanding building aerodynamics and for ensuring accurate wind tunnel test methods. This is especially important in wind tunnel methods that require post-experiment adjustments because approximate wind fields are used, such as the Partial Turbulence Simulation (PTS) approach. Understanding and analyzing these effects enables load adjustments since the PTS method only requires matching the high frequency portions of the upstream spectra of the longitudinal velocity component in model and full-scale. However, the limits for which the PTS method is applicable are unclear in terms of the allowable range of wind field characteristics that can be used in the wind tunnel simulation. To address this, the paper utilizes two nondimensional parameters, one representing the small-scale turbulence energy, ES, and the other the large-scale turbulence energy, EL, to elaborate the aerodynamic effects of turbulence intensity and integral length scales in the upstream wind. The results show that the maximum allowable mismatch ratio of integral length scales and of Jensen numbers between model and full-scale simulations depend on the target small-scale turbulence energy and the maximum allowable deviation of small-scale energy. By quantifying the effects of ES and EL on area-averaged pressure coefficients, the allowable limits are identified for wind tunnel test parameters that lead to negligible differences in the resultant pressure coefficient statistics in regions of separated-reattaching flow on the roof of a low-rise building.

Detection rate of gastrin-releasing peptide receptor (GRPr) targeted tracers for positron emission tomography (PET) imaging in primary prostate cancer: a systematic review and meta-analysis.

The gastrin-releasing peptide receptor (GRPr) has gained recognition as a promising target for both diagnostic and therapeutic applications in a variety of human cancers. This study aims to explore the primary tumor detection capabilities of [68Ga] Ga-GRPr PET imaging, specifically in newly diagnosed intra-prostatic prostate cancer lesions (PCa). Following PRISMA-DTA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses of Diagnostic Test Accuracy Studies) guidelines, a systematic literature search was conducted using the Medline, Embase, Scopus, and Web of Science databases. Data regarding patient characteristics and imaging procedure details-including the type of radiotracer used, administered activity, image acquisition time, scanner modality, criteria, and detection rate of index test-were extracted from the included studies. The pooled patient-and lesion-based detection rates, along with their corresponding 95% confidence intervals (CI), were calculated using a random effects model. The final analysis included 9 studies involving 291 patients and 350 intra-prostatic lesions with [68Ga] Ga-GRPr PET imaging in primary PCa. In per-patient-based analysis of [68Ga] Ga-GRPr PET imaging, the pooled detection rates of overall and patients with Gleason score ≥ 7 were 87.09% (95% CI 74.98-93.82) and 89.01% (95% CI 68.17-96.84), respectively. In per-lesion-based analysis, the pooled detection rate [68Ga] Ga-GRPr PET imaging was 78.54% (95% CI 69.8-85.29). The pooled detection rate mpMRI (multiparametric magnetic resonance imaging) in patient-based analysis was 91.85% (95% CI 80.12-96.92). The difference between the detection rates of the mpMRI and [68Ga] Ga-GRPr PET imaging was not statistically significant (OR 0.90, 95% CI 0.23-3.51). Our findings suggest that [68Ga] Ga-GRPr PET imaging has the potential as a diagnostic target for primary PCa. Future research is needed to determine the effectiveness of [68Ga] Ga-GRPr PET in delivering additional imaging data and guiding therapeutic decisions.

Molecular Classification with Graph ConvolutionalNetworks: Exploring the MUTAG Dataset for Mutagenicity Prediction

This paper presents the implementation of a Graph Convolutional Network (GCN) for the classification of chemical compounds using the MUTAG dataset, which consists of 188 ni- troaromatic compounds labeled according to their mutagenicity. The GCN model leverages the inherent graph structure of molec-ular data to capture and learn from the relationships between atoms and bonds, represented as nodes and edges, respectively. By utilizing three graph convolutional layers followed by a global mean pooling layer, the model effectively aggregates node features to generate meaningful graph-level representations. The model was trained using the Adam optimizer with a learning rate of 0.01, and cross-entropy loss was employed to supervise the classification task. The results demonstrate the efficacy of GCNs in graph classification tasks, with the model achieving a training accuracy of 79.33% and a test accuracy of 76.32%. This study highlights the potential of GCNs in cheminformatics and other domains where graph-structured data is prevalent, paving the way for further exploration and application of advanced graph neural networks in similar tasks.

Automatic defect detection in infrared thermal images of ancient polyptychs based on numerical simulation and a new efficient channel attention mechanism aided Faster R-CNN model

In recent years, the preservation and conservation of ancient cultural heritage necessitate the advancement of sophisticated non-destructive testing methodologies to minimize potential damage to artworks. Therefore, this study aims to develop an advanced method for detecting defects in ancient polyptychs using infrared thermography. The test subjects are two polyptych samples replicating a 14th-century artwork by Pietro Lorenzetti (1280/85–1348) with varied pigments and artificially induced defects. To address these challenges, an automatic defect detection model is proposed, integrating numerical simulation and image processing within the Faster R-CNN architecture, utilizing VGG16 as the backbone network for feature extraction. Meanwhile, the model innovatively incorporates the efficient channel attention mechanism after the feature extraction stage, which significantly improves the feature characterization performance of the model in identifying small defects in ancient polyptychs. During training, numerical simulation is utilized to augment the infrared thermal image dataset, ensuring the accuracy of subsequent experimental sample testing. Empirical results demonstrate a substantial improvement in detection performance, compared with the original Faster R-CNN model, with the average precision at the intersection over union = 0.5 increasing to 87.3% and the average precision for small objects improving to 54.8%. These results highlight the practicality and effectiveness of the model, marking a significant progress in defect detection capability, providing a strong technical guarantee for the continuous conservation of cultural heritage, and offering directions for future studies.

An efficient tracking differentiator based active disturbance rejection control for flight environment simulation system

The flight environment simulation system plays a crucial role in aeroengine testing, replicating real mission pressure and temperature conditions. However, existing systems face challenges in handling strong measurement noises and total disturbances during transient tests, leading to inaccurate simulations and potential actuator damage. To address these gaps, an enhanced tracking differentiator based active disturbance rejection control scheme is proposed and implemented. Our method introduces two key innovations: a discrete-time optimal control law that dynamically adjusts control inputs based on state transition and sampling times, and a phase-advancing method that mitigates filtering phase delays by incorporating derivative signals to predict input trends. These components are integrated into a novel tracking differentiator, enhancing the estimation capability of the extended state observer in active disturbance rejection control. Frequency-domain analysis demonstrates the proposed tracking differentiator's superior filtering performance and phase quality. Simulations of intake environment pressure during aeroengine transient tests reveal that, compared to traditional tracking differentiator-based active disturbance rejection control, the proposed approach significantly reduces both control signal oscillations and setpoint deviations by 75.44% and 53.9%, respectively. These improvements effectively address environment simulation deviations and unnecessary actuator wear, thereby enhancing the accuracy and reliability of aeroengine testing.

Droplet digital PCR for fish pathogen detection and quantification: A systematic review and meta-analysis.

This study provides a comprehensive summary of the findings regarding the application and diagnostic efficacy of droplet digital PCR (ddPCR) in detecting viral and bacterial pathogens in aquaculture. Utilizing a systematic search of four databases up to 6 November 2023, we identified studies where ddPCR was deployed for pathogen detection in aquaculture settings, adhering to Preferred Reporting Items for Systematic Reviews and Meta-analysis of Diagnostic Test Accuracy guidelines. From the collected data, 16 studies retrieved, seven were included in a meta-analysis, encompassing 1121 biological samples from various fish species. The detection limits reported ranged markedly from 0.07 to 34 copies/μL. A direct comparison of the diagnostic performance between ddPCR with quantitative PCR (qPCR) proved challenging due to limited data, thus only a pooled sensitivity analysis was feasible. The results showed a pooled sensitivity of 0.750 (95% confidence interval [CI]: 0.487-0.944) for ddPCR, compared to 0.461 (95% CI: 0.294-0.632) for qPCR, with no statistically significant difference in sensitivity between the two methods (p = .5884). Notably, significant heterogeneity was observed among the studies (I2 = 93%-97%, p < .01), with the year of publication significantly influencing this heterogeneity (p < .001), but not the country of origin (p = .49). No publication bias was detected, and the studies generally exhibited a low risk of bias according to QUADAS-C criteria. While ddPCR and qPCR showed comparable sensitivities in pathogen detection, ddPCR's capability to precisely quantify pathogens without the need for standard curves highlights its potential utility. This characteristic could significantly enhance the accuracy and reliability of pathogen detection in aquaculture.

A deep learning approach for white blood cells image generation and classification using SRGAN and VGG19

The classification of White Blood Cells (WBCs) is crucial for diagnosing diseases, monitoring treatment effectiveness, and understanding how the immune system functions. In this paper, we propose a deep learning approach to classify WBCs using Super Resolution Generative Adversarial Network (SRGAN) and Visual Geometry Group 19 (VGG19). Firstly, microscopic images of WBCs are generated using the SRGAN to obtain more precise and high-resolution images, which are then classified with a pretrained VGG19 classifier. Low-resolution (LR) images are inputted into the generator of SRGAN, and its discriminator compares the High-resolution (HR) image with LR, generating super-resolution images to minimize misclassification risks. A large dataset of 12,447 images containing four classes of WBCs (Eosinophil, Lymphocyte, Monocyte, and Neutrophil) is utilized to train and validate our proposed model. Following extensive experimental analysis, our proposed model achieves a test accuracy of 94.87 %, surpassing traditional Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), Hybrid CNN-RNN models, and other conventional approaches. The generated images of SRGAN overcome challenges associated with misclassification due to the poor resolution of microscopic images, while the use of a pretrained model as a classifier reduces classification complexity. The source code of the entire work is available at https://github.com/Jannatul-Ferdousi/SRGAN_VGG19_WBC.git.