Low Detection Rate Research Articles

Influenza vaccine effectiveness in preventing hospital encounters for laboratory-confirmed infection among Italian adults, 2022/23 season

The effectiveness of seasonal influenza vaccination (SIV) varies from year to year. Interim estimates of vaccine effectiveness (VE) in outpatient settings have suggested that the 2022/23 northern hemisphere SIV was 54 % effective. The main goal of this study was to measure the 2022/23 SIV VE among Italian adults in a hospital setting. The study adopted a retrospective test-negative case-control design and was conducted in a large tertiary hospital (Genoa, Italy) between October 2022 and April 2023. Adult (≥18 years) patients accessing the hospital’s Emergency Department for symptoms ascribable to an acute respiratory infection, for which a reverse-transcription real-time polymerase chain reaction test for the detection of influenza virus was prescribed, were potentially eligible. Of 33,692 referrals assessed, 487 patients were included in the study. A total of 13 % of patients were positive for influenza, most of which (63 %) belonged to the A(H3N2) subtype. SIV VE was 57 % (95 % CI: 11–81 %), 53 % (95 % CI: 2–80 %) and 38 % (95 % CI: −34–74 %) against any influenza, influenza A and A(H3N2), respectively. Although no cases caused by A(H1N1)pdm09 and B strains were observed among vaccinated individuals, VE estimates against the latter were imprecise, owing to their low detection rates. In conclusion, the 2022/23 SIV was moderately effective against hospital encounters for laboratory-confirmed influenza.

A Lightweight Machine Learning-Based Email Spam Detection Model Using Word Frequency Pattern

This Spam emails have become a severe challenge that irritates and consumes recipients' time. On the one hand, existing spam detection techniques have low detection rates and cannot tolerate high-dimensional data. Moreover, due to the machine learning algorithm's effectiveness in identifying mail as solicited or unsolicited, their approaches have become common in spam detection systems. This paper proposes a lightweight machine learning-based spam detection model based on Random Forest (RF) algorithm. According to the empirical results, the proposed model achieved a 97% accuracy on the spambase dataset. Furthermore, the performance of the proposed model was evaluated using standard classification metrics such as Fscore, Recall, Precision, and Accuracy. The comparison of Our model with state-of-the-art works investigated in this paper showed the model performs better, with an improvement of 6% for all metrics.

Frequent detection and genotyping of human rhinovirus in SARS-CoV-2 negative patients; a study from south of Iran.

Human rhinovirus (HRV), a major cause of common cold, was associated to the hospitalization of children and adults. This cross-sectional study aimed to determine the prevalence, and genotype distribution of HRV in the patients with mild to severe respiratory infections who were negative for SARS-Cov-2. Nasopharyngeal swab specimens (n = 356) from the patients aged 29 days to 82 years, received for the respiratory virus detection from January to December 2021, were analyzed for human rhinovirus (HRV) by RT-PCR. As a final step, genotyping was performed on obtained sequences. A total of 37 HRV infections were identified (37/356, 10%). The highest rates of positive HRV tests were observed in February (21.6%), and January (18.9%), compared with June and August (0%). HRV-positive cases mainly appeared in winter. Among the age groups, those 2-<5 years of age had the highest detection rate (21%), however, those >55 years of age had the lowest detection rate (3%). Among HRV-positive samples, 30 (81%) were identified as type HRV-A, 5 (13.5%) as HRV-B, and 2 (5.5%) as HRV-C. Our results suggested that HRV frequency gradually decreased with the age of patients which is more active in Iran, especially in the cold months.

Personalized follow-up of circulating DNA in resected stage III/IV melanoma: PERCIMEL multicentric prospective study protocol

BackgroundWith more than 15,000 new cases /year in France and 2,000 deaths, cutaneous melanoma represents approximately 4% of incidental cancers and 1.2% of cancer related deaths. In locally advanced (stage III) or resectable metastatic (stage IV) melanomas, medical adjuvant treatment is proposed and recent advances had shown the benefit of anti-PD1/PDL1 and anti-CTLA4 immunotherapy as well as anti-BRAF and anti-MEK targeted therapy in BRAF V600 mutated tumors. However, the recurence rate at one year is approximately 30% and justify extensive research of predictive biomarkers. If in metastatic disease, the follow-up of circulating tumor DNA (ctDNA) has been demonstrated, its interest in adjuvant setting remains to be precised, especially because of a lower detection rate. Further, the definition of a molecular response could prove useful to personalized treatment.MethodsPERCIMEL is an open prospective multicentric study executed through collaboration of the Institut de Cancérologie de Lorraine (non-profit comprehensive cancer center) and 6 French university and community hospitals. A total of 165 patients with resected stage III and IV melanoma, eligible to adjuvant imunotherapy or anti-BRAF/MEK kinase inhibitors will be included. The primary endpoint is the presence of ctDNA, 2 to 3 weeks after surgery, defined as mutated ctDNA copy number calculated as the allelic fraction of a clonal mutation relative to total ctDNA. Secondary endpoints are recurrence-free survival, distant metastasis-free survival and specific survival. We will follow ctDNA along treatment, quantitatively through ctDNA mutated copy number variation, qualitatively through the presence of cfDNA and its clonal evolution. Relative and absolute variations of ctDNA during follow-up will be also analyzed.PERCIMEL study aims at provide scientific evidence that ctDNA quantitative and qualitative variations can be used to predict the recurrence of patients with melanoma treated with adjuvant immunotherapy or kinase inhibitors, thus defining the notion of molecular recurrence.

Analysis of SARS-CoV-2 in wastewater for prevalence estimation and investigating clinical diagnostic test biases

Here we analyze SARS-CoV-2 genome copies in Catalonia’s wastewater during the Omicron peak and develop a mathematical model to estimate the number of infections and the temporal relationship between reported and unreported cases. 1-liter samples from 16 wastewater treatment plants were collected and used in a compartmental epidemiological model. The average correlation between genome copies and reported cases was 0.85, with an average delay of 8.8 days. The model estimated that 53% of the population was infected, compared to the 19% reported cases. The under-reporting was highest in November and December 2021. The maximum genome copies shed in feces by an infected individual was estimated to range from 1.4×108 gc/g to 4.4×108 gc/g. Our framework demonstrates the potential of wastewater data as a leading indicator for daily new infections, particularly in contexts with low detection rates. It also serves as a complementary tool for prevalence estimation and offers a general approach for integrating wastewater data into compartmental models.

Single-Point Nail Sampling to Diagnose Onychomycosis Caused by Non-Dermatophyte Molds: Utility of Polymerase Chain Reaction (PCR) and Histopathology

The three most commonly used methods for diagnosing non-dermatophyte mold (NDM) onychomycosis are culture, polymerase chain reaction (PCR), and histopathology. Toenail samples from 512 patients (1 sample/patient) with suspected onychomycosis were examined using all three diagnostic tests. A statistically significant association was found between PCR and histopathology results, as well as between fungal culture and histopathology results. All PCR-positive and culture-positive dermatophyte samples were confirmed by histopathology. However, 15/116 (12.9%) of culture-positive NDM samples had negative histopathology results, while all PCR-positive NDM samples were confirmed by histopathology. The overall rate of dermatophyte detection was higher using PCR compared to culture (38.9% vs. 11.7%); the lower rate of NDM detection by PCR (11.7% vs. 38.9%) could be attributed to the restriction of the assay design to seven pre-selected targets. When repeat sampling in the clinic is not possible, a combination of NDM detection by PCR and positive histopathology of hyphae may be a proxy for NDM infection, particularly where the NDM occurs without a concomitant dermatophyte. There was a high degree of correlation between negative PCR and negative histopathology. A negative PCR result with negative histopathology findings may be a reliable proxy for the diagnosis of non-fungal dystrophy.

Learning discriminative feature representation with pixel-level supervision for forest smoke recognition

Existing vision-based smoke recognition methods still face the issues of low detection rates and high false alarm rates in complex scenes. One reason is that they label light smoke and heavy smoke as the same value, which ignores the differences in multiple attribute information involved in the smoke imaging process. To solve this issue, this paper presents a pixel-level supervision neural network (PSNet) to learn discriminative feature representations for forest smoke recognition. First, the pixel-level supervision information, including the background component, smoke component, fusion ratio, and class information, is cooperatively considered to effectively guide the model training process. To avoid negative transfer caused by the asynchronous optimization of shared layer parameters and achieve synchronous minimization of each loss term, a regularization term based on the smoke imaging principle and a weight dynamic updating method are proposed to balance the weight coefficients of different loss terms. Second, a detail-difference-aware module (DDAM) based on a detail-difference-aware block (DDAB) and a spatial attention block (SAB) is proposed to distinguish smoke and smoke-like targets by fusing xy-shared convolution and z-shared convolution, which adaptively allocates the weights over different positions to prioritize the most informative visual elements in the spatial domain. Third, an attention-based feature separation module (AFSM) is proposed to relieve mutual interference in extracting background features and smoke features by designing component interaction attention (CIA), background component attention (BCA), smoke component attention (SCA), and enhanced residual blocks (ERBs), which can guide the interaction and separation process of background information and smoke information to enhance the discriminative spatial features and suppress interference features. ERB effectively eliminates noise and enhances smoke edge information based on median filters. Finally, to further enhance the feature representation capability, a multiconnection aggregation method (MCAM) is proposed by fully aggregating local and global features simultaneously. Extensive experiments show that our method achieves better performance than existing smoke recognition methods.Extensive experiments show that our PSNet achieves better performance than existing smoke recognition methods. For smoke recognition, our PSNet achieves a 96.95% detection rate, 3.02% false alarm rate, and 0.9694 F1-score. The average calculation time for each image is only 0.0195. For smoke component separation, our PSNet also achieves 0.0014 on evaluation criteria mean square error between predicted smoke component images and labelled smoke component images. These key experimental results are better than those of previous methods.

Obstrucción congénita de la unión ureteropiélica. Estudio de casos en un centro de segundo nivel

Objective: To evaluate the incidence, clinical characteristics, diagnosis, and treatment of congenital ureteropelvic junction obstruction in pediatric patients (CUPBO) at the Hospital Infantil de Morelia. Material and methods: This was a retrospective, observational, descriptive, and cross-sectional study of pediatric patients diagnosed with CUPBO and treated in the urology service of a second-level care medical center from January 2018 to December 2021. Results: We reviewed 53 patients with prenatal hydronephrosis who underwent postnatal follow-up, of whom only 6 patients diagnosed with congenital ureteropelvic junction obstruction (CUPBO) were identified prenatally. Seven patients were diagnosed with hydronephrosis postnatally and subsequently diagnosed with CUPBO, resulting in a total sample of 13 diagnosed patients. The incidence of CUPBO was 3.4 patients per 100 first visits to the Pediatric Urology Service. Of the cases, 92.31% were male, and the most frequent clinical manifestations were skin infections and localized pain. Surgery was performed in all cases, with Anderson-Hynes pyeloplasty being the most frequent procedure. Complications included urinary tract infections, wound infections, and fistula, but most patients had satisfactory results with a reduction of hydronephrosis after surgery. Findings or conclusions: Our study found a low rate of early detection of CUPBO in patients with prenatal hydronephrosis follow-up. We recommend an intentional search for urologic alterations in structural ultrasound scans of prenatal control. We emphasize the importance of an accurate diagnosis and timely treatment of CUPBO to prevent renal complications.

Analyzing molecular typing and clinical application of immunogenic cell death-related genes in hepatocellular carcinoma

BackgroundHepatocellular carcinoma (HCC) is considered one of the most common cancers, characterized by low early detection and high mortality rates, and is a global health challenge. Immunogenic cell death (ICD) is defined as a specific type of regulated cell death (RCD) capable of reshaping the tumor immune microenvironment by releasing danger signals that trigger immune responses, which would contribute to immunotherapy.MethodsThe ICD gene sets were collected from the literature. We collected expression data and clinical information from public databases for the HCC samples in our study. Data processing and mapping were performed using R software to analyze the differences in biological characteristics between different subgroups. The expression of the ICD representative gene in clinical specimens was assessed by immunohistochemistry, and the role of the representative gene in HCC was evaluated by various in vitro assays, including qRT-PCR, colony formation, and CCK8 assay. Lasso-Cox regression was used to screen prognosis-related genes, and an ICD-related risk model (ICDRM) was constructed. To improve the clinical value of ICDRM, Nomograms and calibration curves were created to predict survival probabilities. Finally, the critical gene of ICDRM was further investigated through pan-cancer analysis and single-cell analysis.ResultsWe identified two ICD clusters that differed significantly in terms of survival, biological function, and immune infiltration. As well as assessing the immune microenvironment of tumors in HCC patients, we demonstrate that ICDRM can differentiate ICD clusters and predict the prognosis and effectiveness of therapy. High-risk subpopulations are characterized by high TMB, suppressed immunity, and poor survival and response to immunotherapy, whereas the opposite is true for low-risk subpopulations.ConclusionsThis study reveals the potential impact of ICDRM on the tumor microenvironment (TME), immune infiltration, and prognosis of HCC patients, but also a potential tool for predicting prognosis.

Efficient e-mail spam filtering approach combining Logistic Regression model and Orthogonal Atomic Orbital Search algorithm

Phishing emails called spam have created a need for reliable and intelligent spam filters. Machine-learning techniques are effective, but current methods such as Logistic Regression (LR), Support Vector Machine (SVM), Decision Trees (DT), and Naive Bayes (NB) sometimes produce low detection rates and struggle with large amounts of data. Motivated by these concerns, we propose an efficient spam-filtering approach, OAOS-LR, combining an improved Atomic Orbital Search (AOS) algorithm with an LR classification model. To remove the deficiency of low detection rate produced by the standard LR method due to the utilization of the gradient descent technique, we train it with our proposed OAOS approach, which uses AOS and Orthogonal learning to enhance the search capabilities of the conventional algorithm. In the experimental study, we first evaluated the performance of OAOS over the IEEE Congress on Evolutionary Computation (CEC’20) benchmarks against five different metaheuristics to prove its effectiveness in improving its convergence rate and reducing the probability of falling in local optima. After that, the proposed technique LR-OAOS was applied to the spam filtering problem using CSDMC2010 and Enron datasets and tested against the most recent machine learning and metaheuristic approaches using standard statistical measures and plots. OAOS-LR significantly outperformed other methods with an average F1-score success rate of 95.45% and 96.30% on CSDMC2010, and 74.80% and 78.33% on Enron, respectively with the number of feature spaces equal to 500 and 1000.

Biopsy Assessment of Oncologic Control 3 Years Following Primary Partial Gland Cryoablation: A Prospective Cohort Study of Men With Intermediate-risk Prostate Cancer.

We evaluated 3-year oncologic outcomes following primary partial gland cryoablation. Men with unilateral intermediate-risk prostate cancer undergoing primary partial gland cryoablation since March 2017 enrolled in a prospective outcome registry. The postablation protocol for all men included surveillance prostate biopsy at 2 years postablation and reflex prostate biopsy for cases with high suspicion of recurrence (eg, progressive rise in PSA). Recurrence of clinically significant prostate cancer was defined as any Gleason grade group ≥2 disease on postablation biopsy. Freedom from failure represented no whole gland salvage treatment, metastatic prostate cancer, or prostate cancer mortality. Freedom from recurrence and freedom from failure were characterized using nonparametric maximum likelihood estimators. A total of 132 men had at least 24 months of follow-up data. Biopsies identified clinically significant prostate cancer in 12 men. At 36 months, model-estimated rates of freedom from recurrence of in-field, out-of-field, and overall clinically significant cancer were 97% (95% CI: 92-100), 87% (95% CI: 80-94), and 86% (95% CI: 78-93), respectively. The model-estimated proportion with freedom from failure at 36 months was 97% (95% CI: 93-100). The low in-field cancer detection rate at 3 years indicates successful ablation of localized cancers. Conversely, our observed out-of-field detection rate highlights the need for continued surveillance following partial gland cryoablation. Many of these recurrences exhibited very low volume of clinically significant disease below the detection threshold of multiparametric MRI, suggesting a limited role for multiparametric MRI in detecting clinically significant recurrences at 2 years. These findings emphasize the need for long-term surveillance and identification of predictors of clinically significant prostate cancer recurrences to guide biopsy timing.

The Effelsberg survey of FU Orionis and EX Lupi objects

Context. FU Orionis (FUor) and EX Lupi (EXor) type objects are two groups of peculiar and rare pre-main sequence low-mass stars that are undergoing powerful accretion outbursts during their early stellar evolution. Although water masers are widespread in star-forming regions and serve as powerful probes of mass accretion and ejection on small scales, little is known about the prevalence of water masers toward FUors and EXors. Aims. We aim to perform the first systematic search for the 22.2 GHz water maser line in FUors and EXors in order to determine its overall incidence in these eruptive variables and to facilitate high angular-resolution follow-up observations. Methods. We used the Effelsberg 100-m radio telescope to observe the H2O (616 – 523) transition at 22.2GHz toward a sample of 51 eruptive young stellar objects. Results. We detected a total of five water masers in our survey; three are associated with eruptive stars, equivalent to a detection rate of ~6% for our sample of eruptive sources. These detections include one EXor, V512 Per (also known as SVS 13 or SVS 13A), and two FUors, Z CMa, and HH 354 IRS. This is the first reported detection of water maser emission towards HH 354 IRS. We also detected water maser emission in our pointing towards the FUor binary RNO 1B/1C, which most likely originates from the nearby deeply embedded source IRAS 00338+6312 (~4″ from RNO 1B/1C). Emission was also detected from H2O(B), (also known as SVS 13C), a Class 0 source ~30″ from the EXor V512 Per. The peak flux density of H2O(B) in our observations, 498.7 Jy, is the highest observed to date, indicating that we have serendipitously detected a water maser flare in this source. In addition to the two non-eruptive Class 0 sources (IRAS 00338+6312 and H2O(B)/SVS 13C), we detected maser emission towards one Class 0/I (HH 354 IRS) and two Class I (V512 Per/SVS 13A and Z CMa) eruptive stars. Conclusions. Despite the low detection rate, we demonstrate the presence of 22.2 GHz water maser emission in both FUor and EXor systems, paving the way to radio interferometric observations to study the environments of these eruptive stars on small scales. Comparisons of our data with historical observations spanning several decades suggest that multiple water maser flares have occurred in both V512 Per and H2O(B).

The MOSDEF-LRIS survey: detection of inflowing gas towards three star-forming galaxies at z ∼ 2

ABSTRACT We report on the discovery of cool gas inflows towards three star-forming galaxies at &amp;lt;z&amp;gt; ∼ 2.30. Analysis of Keck Low-Resolution Imaging Spectrometer spectroscopy reveals redshifted low-ionization interstellar (LIS) metal absorption lines with centroid velocities of 60–130 km s−1. These inflows represent some of the most robust detections of inflowing gas into isolated, star-forming galaxies at high redshift. Our analysis suggests that the inflows are due to recycling metal-enriched gas from previous ejections. Comparisons between the galaxies with inflows and a larger parent sample of 131 objects indicate that galaxies with detected inflows may have higher specific star formation rates (sSFRs) and star-formation-rate surface densities (ΣSFR). However, when additional galaxies without robustly detected inflows based on centroid velocity but whose LIS absorption line profiles indicate large red-wing velocities are considered, galaxies with inflows do not show unique properties relative to those lacking inflows. Additionally, we calculate the covering fraction of cool inflowing gas as a function of red-wing inflow velocity, finding an enhancement in high-sSFR binned galaxies, likely due to an increase in the amount of recycling gas. Together, these results suggest that the low detection rate of galaxies with cool inflows is primarily related to the viewing angle rather than the physical properties of the galaxies.

Clinical characterization of MYC family proto-oncogene amplification in solid tumors from Chinese patients.

e15140 Background: The dysregulation of the MYC family oncogenes ( c-MYC, MYCN and MYCL) play critical roles in tumorigenesis, prognosis and immune escape. MYC inactivation can result in sustained tumour regression and many therapeutic agents that directly target MYC are under development. MYC signaling is associated with tumor cell PD-L1, overall immune cell infiltration. Herein, we explore MYC family proto-oncogene amplification profiles and clinical characterization in chinese solid tumors. Methods: This research comprehensively characterized gene mutations by next-generation sequencing (NGS) in 23990 chinese solid tumors tissues to reveal the prevalence of MYC family proto-oncogene amplification(MYC AMP) and the association with Tumor mutational burden(TMB) and microsatellite instability(MSI). Results: The prevalence of MYC AMP (copy number, CN≥5) in the cohort was 2.1% (504/23,990), in which ovarian cancer (7.6%, 22/289) showed the highest prevalence, followed by breast cancer (6.1%, 26/429), esophagus cancer (5.9%, 17/287). Only one glioma patient(pt) carried co-amplification of MYCN and c-MYC. In 504 MYC AMP pts, c-MYC AMP accounted for 93.7%(472/504), MYCN and MYCL AMP accounted for 6.5%(33/504) in total. The CN was significantly higher in MYCN and MYCL AMP pts than c-MYC AMP pts (22.9 vs 7.6, p &lt; 0.0001). MSI-H showed a lower detection rate in MYC AMP pts other than Non-MYC AMP pts (0% vs 1.2%, p &lt; 0.05). The proportion of TMB-L in MYC AMP pts was similar to Non-MYC AMP pts (90.9% vs. 90.1%, p &gt; 0.05). Conclusions: In totally, 2.1% of chinese solid tumor pts had MYC high level AMP, mainly c-MYC Amp. The CN was higer in MYCN/ MYCL AMP pts than c-MYC AMP pts. In addition, MYC AMP pts tended to have MSS and TMB-L, suggesting that MYC may be a novel target for tumor immunotherapy. MYC inhibitor combines with immunotherapy may be an important direction for the treatment of MYC AMP pts.

AYOLOv5: Improved YOLOv5 based on attention mechanism for blood cell detection

A crucial component of biological study and clinical trials is the microscopic examination of cells and tissues. Although tremendous progress has been made in computer-assisted microscopy cell detection techniques, especially recently based on deep learning, it still needs to be determined to accurately identify cell targets in the presence of dense and complex cell distribution. In this research, an improved YOLOv5 (AYOLOv5) based on the attention mechanism is suggested to address the issue of the low recognition rate of cell detection caused by this circumstance. Based on YOLOv5, the attention mechanism is introduced to improve the convolutional neural network's features in areas of the picture with a high density of features. The convolutional block attention module (CBAM) and the transformer encoder block are used in this study to develop the attention mechanism. The YOLOv5 integrated convolutional block attention module increases the weight of cell-dense regions in blood cell pictures and aids the network's ability to resist information other than cells. Additionally, the transformer block is introduced to YOLOv5′s processing of upper and lower feature data to improve the network's capacity to gather details about various cell properties, enabling AYOLOv5 to recognize and distinguish blood cells in cell-dense areas. The experiment was done on the dataset BCCD, and the mAP results for cell detection reached 93.3%, better than previously discovered. Also, the validation set's average recognition accuracy increased from 89% to 98%. The experimental results demonstrated that the suggested AYOLOv5 could extract the cells' feature information more effectively, considerably improving the cell pictures and recognition performance.

Bisphenol and phthalate exposure during pregnancy and the development of childhood lung function and asthma. The Generation R Study

BackgroundFetal exposure to bisphenols and phthalates may lead to alterations in the respiratory and immune system development in children, and to adverse respiratory health. AimTo study the associations of fetal bisphenols and phthalates exposure with lung function and asthma at age 13 years. Study design and methodsThis study among 1020 children was embedded in a population-based prospective cohort study. We measured maternal urine bisphenol and phthalate concentrations in the first, second and third trimester of pregnancy, and lung function by spirometry and asthma by questionnaires at age 13 years. Multivariable linear and logistic regression models were applied. ResultsMaternal urine bisphenol and phthalate concentrations averaged during pregnancy were not associated with childhood lung function or asthma. Associations of maternal urine bisphenol and phthalate concentrations in specific trimesters with respiratory outcomes showed that one interquartile range increase in the natural log-transformed maternal urine mono-isobutyl phthalate concentration in the second trimester was associated with a higher FEV1/FVC, but not with asthma, accounting for confounders and multiple-testing correction. Although there were associations of higher second trimester bisphenol S with a lower FVC and FEV1 in boys and girls, and of higher first trimester bisphenol S with a decreased risk of asthma in boys and an increased risk of asthma in girls, these results did not remain significant after correction for multiple testing. Results were not modified by maternal history of asthma or atopy. ConclusionsMaternal urine bisphenol and phthalate concentrations averaged or in specific trimesters during pregnancy were not strongly associated with childhood lung function and asthma at age 13 years. BPS, as a BPA substitute, tended to be associated with impaired lung function and altered risk of asthma, partly sex-dependent, but its strength was limited by a relatively low detection rate and should be queried in contemporary cohorts.

Machine Learning-Based Adaptive Synthetic Sampling Technique for Intrusion Detection

Traditional firewalls and data encryption techniques can no longer match the demands of current IoT network security due to the rising amount and variety of network threats. In order to manage IoT network risks, intrusion detection solutions have been advised. Even though machine learning (ML) helps the widely used intrusion detection techniques currently in use, these algorithms struggle with low detection rates and the requirement for extensive feature engineering. The deep learning model for IoT network intrusion detection is a method for traffic anomaly detection that is suggested by this study. To extract the sequence properties of data flow through a CNN, it combines an attention mechanism with a Long Short Term Memory (LSTM) network. This method uses adaptive synthetic sampling (ADASYN) to increase the size of minority-class samples. The proposed models demonstrated acceptable precision and recall for each class when used for binary-class classification, proving their stability and capacity to identify all classes correctly. The MLP classifier’s accuracy, precision, recall, and F1 value were 87%, 89%, 87%, and 89%, respectively, with an AUC score of 0.88. Overall, the proposed models performed well. The attack and all-class models exhibited good AUCs and macro metrics, the same as the proposed MLP classifier, which had an F1 score of 83% and an AUC score of 0.94. Additionally, it trained the MLP classifier and integrated the ADAM optimizer and category cross-entropy loss function for all-class classification. With an AUC value of 94%, it possessed 84% accuracy, 87% precision, 84% recall, and an 83% F1 score. A further indication of the hybrid model’s ability to combine the benefits of both models to improve overall performance was that it regularly outperformed the MLP model. This model’s accuracy and F1 score are better than those of earlier comparable algorithms, according to experimental results using the publicly accessible benchmark dataset for network intrusion detection (NSL–KDD).

Poor performance of malaria rapid diagnostic tests for the detection of Plasmodium malariae among returned international travellers in China

BackgroundMalaria is a worldwide infectious disease. For countries that have achieved malaria elimination, the prevention of re-establishment due to infections in returned travellers has become important. The accurate and timely diagnosis of malaria is the key in preventing re-establishment, and malaria rapid diagnostic tests (RDTs) are frequently used due to their convenience. However, the RDT performance in Plasmodium malariae (P. malariae) infection diagnosis remains unknown.MethodsThis study analysed epidemiological features and diagnosis patterns of imported P. malariae cases from 2013 to 2020 in Jiangsu Province and evaluated the sensitivity of four parasite enzyme lactate dehydrogenase (pLDH)-targeting RDTs (Wondfo, SD BIONLINE, CareStart and BioPerfectus) and one aldolase-targeting RDT(BinaxNOW) for P. malariae detection. Furthermore, influential factors were investigated, including parasitaemia load, pLDH concentration and target gene polymorphisms.ResultsThe median duration from symptom onset to diagnosis among patients with P. malariae infection was 3 days, which was longer than that with Plasmodium falciparum (P. falciparum) infection. The RDTs had a low detection rate (39/69, 56.5%) among P. malariae cases. All tested RDT brands had poor performance in P. malariae detection. All the brands except the worst-performing SD BIOLINE, achieved 75% sensitivity only when the parasite density was higher than 5000 parasites/μL. Both pLDH and aldolase showed relatively conserved and low gene polymorphism rates.ConclusionsThe diagnosis of imported P. malariae cases was delayed. The RDTs had poor performance in P. malariae diagnosis and may threaten the prevention of malaria re-establishment from returned travellers. The improved RDTs or nucleic acid tests for P. malariae cases are urgently needed for the detection of imported cases in the future.

Detection and quantification of cotton trichomes by deep learning algorithm

Cotton is one of the most economically important agricultural crops with the harvest of cotton fibers. Trichomes are initiated from the plant epidermis, which is responsible for plant resistance to many abiotic and biotic stresses. Cotton fibers share similar underlying developmental molecular mechanisms with trichomes on the stem and leaf margins of cotton. Compared with fibers, it is easier to observe the trichome phenotype of cotton stem and leaf margins. However, because of the variability in coloration and the high occlusion of the trichomes, counting the density and observing the distribution of trichomes is still highly laborious. To help resolve this problem, the present research provides a solution based on deep learning in the screening of cotton trichomes, which makes it possible to monitor the phenotype of the cotton efficiently. With a fluorescent microscopy imaging procedure, we clearly observed that there are two types of trichome phenotype inGossypium hirsutum(‘TM-1′), which are singled and clustered trichomes. The clustered trichome originates from two or more adjacent epidermal cells with no branches. The TM1 trichome was used to create a collection of singled and clustered trichome pictures. Following data production, we evaluated the effectiveness of several you only look once (YOLO) model iterations in the context of identifying singled and clustered trichomes, respectively. Results show that the YOLOv5s model outperforms YOLOv3 and YOLOv4 in average accuracy, recall, accuracy, and F1 score. We utilized the clustered trichome data set to assess the performance of YOLOv5s and Mask R-CNN in order to more precisely detect the real trichome distribution with simultaneous detection of single and clustered trichomes. The obtained findings show that both models may be utilized to concurrently identify singled and clustered trichomes since the discrepancy between the detected values of the two models and the real values is quite minimal. Mask R-CNN has a high IOU value, whereas YOLOv5s has a low false-positive detection rate and fast detection speed. Our models and datasets can be used as a starting point to measure the trichome trait in other cotton species and establish a cotton trichome detection and counting system based on deep learning technology.

The Efficacy of AI-Assisted CADe Systems in Colorectal Polyps and Adenomas during Colonoscopies: A Systematic Review

Colorectal polyps and adenomas are a significant health concern that can lead to colorectal cancer. Detecting polyps and adenomas early is important to remove them in a timely manner, reducing the risk of cancer. Artificial intelligence (AI) could assist in identifying polyps and adenomas more efficiently. This study will investigate the potential benefits of AI technology in detecting colorectal adenomas, using AI-assisted colonoscopy diagnostic systems. The goal is to determine if there are differences in detection rates using AI compared to standard colonoscopy methods. PRISMA Statement 2020 guidelines were adhered to in this systematic review. Three databases were searched including PubMed/MEDLINE, Embase and Web of Science through February 23, 2023. A combination of the following keywords was used: AI, Artificial, Intelligence, Colorectal, Polyp, Adenoma. The adenoma detection rate (ADR) and polyp detection rate (PDR) were additionally meta-analyzed for proportions, and reported as odds ratio (OR), applying 95% confidence intervals (CI). A total of 9 randomized clinical trials were included in this review, pooling in 6981 participants. The findings were presented systematically, based on the design, technology used, effects on ADR and PDR, and key conclusions. The meta-analytical findings were reported as follows. On applying a random-effects model, the odds ratio (OR) for ADR was 1.532 (95% CI=1.125-2.087), with significant detection favoring CADe systems (P=0.0068). Similarly, when assessing the chance of diagnosing colorectal polyps with CADe systems as an add-on compared to standard colonoscopy alone, the odds ratio (OR) was 1.893 (95% CI=1.687-2.125, P&lt;0.0001). This systematic review found that real-time computer-aided detection during colonoscopy improved the detection of adenomas and polyps. The use of the system is feasible and safe, and future studies should focus on improving colonoscopy quality indicators including the most widely used– ADR. Larger randomized controlled studies are needed to evaluate the impact of this system on adenoma and serrated polyp detection and its effect on endoscopists with lower detection rates.