A cardiologist-like computer-aided interpretation framework to improve arrhythmia diagnosis from imbalanced training datasets.

Arrhythmias can pose a significant threat to cardiac health, potentially leading to serious consequences such as stroke, heart failure, cardiac arrest, shock, and sudden death. In computer-aided electrocardiogram interpretation systems, the inclusion of certain classes of arrhythmias, which we term "aggressive" or "bullying," can lead to the underdiagnosis of other "vulnerable" classes. To address this issue, a method for arrhythmia diagnosis is proposed in this study. This method combines morphological-characteristic-based waveform clustering with Bayesian theory, drawing inspiration from the diagnostic reasoning of experienced cardiologists. The proposed method achieved optimal performance in macro-recall and macro-precision through hyperparameter optimization, including spliced heartbeats and clusters. In addition, with increasing bullying by aggressive arrhythmias, our model obtained the highest average recall and the lowest average drop in recall on the nine vulnerable arrhythmias. Furthermore, the maximum cluster characteristics were found to be consistent with established arrhythmia diagnostic criteria, lending interpretability to the proposed method.

MRI-based brain age prediction model for children under 3 years old using deep residual network.

Early identification and intervention of abnormal brain development individual subjects are of great significance, especially during the earliest and most active stage of brain development in children aged under 3. Neuroimage-based brain's biological age has been associated with health, ability, and remaining life. However, the existing brain age prediction models based on neuroimage are predominantly adult-oriented. Here, we collected 658 T1-weighted MRI scans from 0 to 3 years old healthy controls and developed an accurate brain age prediction model for young children using deep learning techniques with high accuracy in capturing age-related changes. The performance of the deep learning-based model is comparable to that of the SVR-based model, showcasing remarkable precision and yielding a noteworthy correlation of 91% between the predicted brain age and the chronological age. Our results demonstrate the accuracy of convolutional neural network (CNN) brain-predicted age using raw T1-weighted MRI data with minimum preprocessing necessary. We also applied our model to children with low birth weight, premature delivery history, autism, and ADHD, and discovered that the brain age was delayed in children with extremely low birth weight (less than 1000 g) while ADHD may cause accelerated aging of the brain. Our child-specific brain age prediction model can be a valuable quantitative tool to detect abnormal brain development and can be helpful in the early identification and intervention of age-related brain disorders.

A multicenter study on two-stage transfer learning model for duct-dependent CHDs screening in fetal echocardiography.

Duct-dependent congenital heart diseases (CHDs) are a serious form of CHD with a low detection rate, especially in underdeveloped countries and areas. Although existing studies have developed models for fetal heart structure identification, there is a lack of comprehensive evaluation of the long axis of the aorta. In this study, a total of 6698 images and 48 videos are collected to develop and test a two-stage deep transfer learning model named DDCHD-DenseNet for screening critical duct-dependent CHDs. The model achieves a sensitivity of 0.973, 0.843, 0.769, and 0.759, and a specificity of 0.985, 0.967, 0.956, and 0.759, respectively, on the four multicenter test sets. It is expected to be employed as a potential automatic screening tool for hierarchical care and computer-aided diagnosis. Our two-stage strategy effectively improves the robustness of the model and can be extended to screen for other fetal heart development defects.

In silico epitope prediction and evolutionary analysis reveals capsid mutation patterns for enterovirus B.

Enterovirus B (EVB) is a common species of enterovirus, mainly consisting of Echovirus (Echo) and Coxsackievirus B (CVB). The population is generally susceptible to EVB, especially among children. Since the 21st century, EVB has been widely prevalent worldwide, and can cause serious diseases, such as viral meningitis, myocarditis, and neonatal sepsis. By using cryo-electron microscopy, the three-dimensional (3D) structures of EVB and their uncoating receptors (FcRn and CAR) have been determined, laying the foundation for the study of viral pathogenesis and therapeutic antibodies. A limited number of epitopes bound to neutralizing antibodies have also been determined. It is unclear whether additional epitopes are present or whether epitope mutations play a key role in molecular evolutionary history and epidemics, as in influenza and SARS-CoV-2. In the current study, the conformational epitopes of six representative EVB serotypes (E6, E11, E30, CVB1, CVB3 and CVB5) were systematically predicted by bioinformatics-based epitope prediction algorithm. We found that their epitopes were distributed into three clusters, where the VP1 BC loop, C-terminus and VP2 EF loop were the main regions of EVB epitopes. Among them, the VP1 BC loop and VP2 EF loop may be the key epitope regions that determined the use of the uncoating receptors. Further molecular evolution analysis based on the VP1 and genome sequences showed that the VP1 C-terminus and VP2 EF loop, as well as a potential "breathing epitope" VP1 N-terminus, were common mutation hotspot regions, suggesting that the emergence of evolutionary clades was driven by epitope mutations. Finally, footprints showed mutations were located on or near epitopes, while mutations on the receptor binding sites were rare. This suggested that EVB promotes viral epidemics by breaking the immune barrier through epitope mutations, but the mutations avoided the receptor binding sites. The bioinformatics study of EVB epitopes may provide important information for the monitoring and early warning of EVB epidemics and developing therapeutic antibodies.

An Innovative Three-Stage Model for Prenatal Genetic Disorder Detection Based on Region-of-Interest in Fetal Ultrasound.

A global survey has revealed that genetic syndromes affect approximately 8% of the population, but most genetic diagnoses are typically made after birth. Facial deformities are commonly associated with chromosomal disorders. Prenatal diagnosis through ultrasound imaging is vital for identifying abnormal fetal facial features. However, this approach faces challenges such as inconsistent diagnostic criteria and limited coverage. To address this gap, we have developed FGDS, a three-stage model that utilizes fetal ultrasound images to detect genetic disorders. Our model was trained on a dataset of 2554 images. Specifically, FGDS employs object detection technology to extract key regions and integrates disease information from each region through ensemble learning. Experimental results demonstrate that FGDS accurately recognizes the anatomical structure of the fetal face, achieving an average precision of 0.988 across all classes. In the internal test set, FGDS achieves a sensitivity of 0.753 and a specificity of 0.889. Moreover, in the external test set, FGDS outperforms mainstream deep learning models with a sensitivity of 0.768 and a specificity of 0.837. This study highlights the potential of our proposed three-stage ensemble learning model for screening fetal genetic disorders. It showcases the model's ability to enhance detection rates in clinical practice and alleviate the burden on medical professionals.

A Deep-Learning-Based Method Can Detect Both Common and Rare Genetic Disorders in Fetal Ultrasound.

A global survey indicates that genetic syndromes affect approximately 8% of the population, but most genetic diagnoses can only be performed after babies are born. Abnormal facial characteristics have been identified in various genetic diseases; however, current facial identification technologies cannot be applied to prenatal diagnosis. We developed Pgds-ResNet, a fully automated prenatal screening algorithm based on deep neural networks, to detect high-risk fetuses affected by a variety of genetic diseases. In screening for Trisomy 21, Trisomy 18, Trisomy 13, and rare genetic diseases, Pgds-ResNet achieved sensitivities of 0.83, 0.92, 0.75, and 0.96, and specificities of 0.94, 0.93, 0.95, and 0.92, respectively. As shown in heatmaps, the abnormalities detected by Pgds-ResNet are consistent with clinical reports. In a comparative experiment, the performance of Pgds-ResNet is comparable to that of experienced sonographers. This fetal genetic screening technology offers an opportunity for early risk assessment and presents a non-invasive, affordable, and complementary method to identify high-risk fetuses affected by genetic diseases. Additionally, it has the capability to screen for certain rare genetic conditions, thereby enhancing the clinic's detection rate.

The Two-Stage Ensemble Learning Model Based on Aggregated Facial Features in Screening for Fetal Genetic Diseases.

With the advancement of medicine, more and more researchers have turned their attention to the study of fetal genetic diseases in recent years. However, it is still a challenge to detect genetic diseases in the fetus, especially in an area lacking access to healthcare. The existing research primarily focuses on using teenagers' or adults' face information to screen for genetic diseases, but there are no relevant directions on disease detection using fetal facial information. To fill the vacancy, we designed a two-stage ensemble learning model based on sonography, Fgds-EL, to identify genetic diseases with 932 images. Concretely speaking, we use aggregated information of facial regions to detect anomalies, such as the jaw, frontal bone, and nasal bone areas. Our experiments show that our model yields a sensitivity of 0.92 and a specificity of 0.97 in the test set, on par with the senior sonographer, and outperforming other popular deep learning algorithms. Moreover, our model has the potential to be an effective noninvasive screening tool for the early screening of genetic diseases in the fetus.

Case Report: Early diagnosis of lethal multiple pterygium syndrome with micrognathia: Two novel mutations in the CHRND gene.

Lethal multiple pterygium syndrome (LMPS) is a rare disease with genetic and phenotypic heterogeneity and is inherited in an autosomal recessive (AR) pattern. Here, we have presented clinically significant results describing two novel mutations of CHRND gene: NM_000751.2: c.1006C>T p.(Arg336Ter) and NM_000751.2:c.973_975delGTG p.(Val325del), and measurement of the facial angle for determining micrognathia by prenatal diagnosis in the first trimester of pregnancy for a Lethal multiple pterygium syndrome case. In conclusion, this report complements the spectrum of genetic variants and phenotype of Lethal multiple pterygium syndrome and provides reliable recommendation for the counseling of future pregnancies in families with the disease.

PD-L1 negatively regulates antifungal immunity by inhibiting neutrophil release from bone marrow.

Programmed death ligand 1 (PD-L1) has been shown to be inducibly expressed on neutrophils to suppress host immunity during polymicrobial sepsis, virus and parasite infections. However, the role of PD-L1 on neutrophil-mediated antifungal immunity remains wholly unknown. Here, we show that the expression of PD-L1 on murine and human neutrophils was upregulated upon the engagement of C-type lectin receptor Dectin-1 with its ligand ß-glucans, exposed on fungal pathogen Candida albicans yeast. Moreover, ß-glucan stimulation induced PD-L1 translocation into nucleus to regulate the production of chemokines CXCL1 and CXCL2, which control neutrophil mobilization. Importantly, C. albicans infection-induced expression of PD-L1 leads to neutrophil accumulation in bone marrow, through mediating their autocrine secretion of CXCL1/2. Furthermore, neutrophil-specific deficiency of PD-L1 impaired CXCL1/2 secretion, which promoted neutrophil migration from bone marrow into the peripheral circulation, thereby conferring host resistance to C. albicans infection. Finally, either PD-L1 blockade or pharmacological inhibition of PD-L1 expression significantly increased neutrophil release from bone marrow to enhance host antifungal immunity. Our data together indicate that activation of Dectin-1/PD-L1 cascade by ß-glucans inhibits neutrophil release from bone marrow reserve, contributing to the negative regulation of antifungal innate immunity, which functions as a potent immunotherapeutic target against life-threatening fungi infections.

Digital Analysis of Smart Registration Methods for Magnetic Resonance Images in Public Healthcare.

Brain development and atrophy accompany people's life. Brain development diseases, such as autism and Alzheimer's disease, affect a large part of the population. Analyzing brain development is very important in public healthcare, and image registration is essential in medical brain image analysis. Many previous studies investigate registration accuracy by the "ground truth" dataset, marker-based similarity calculation, and expert check to find the best registration algorithms. But the evaluation of image registration technology only at the accuracy level is not comprehensive. Here, we compare the performance of three publicly available registration techniques in brain magnetic resonance imaging (MRI) analysis based on some key features widely used in previous MRI studies for classification and detection tasks. According to the analysis results, SPM12 has a stable speed and success rate, and it always works as a guiding tool for newcomers to medical image analysis. It can preserve maximum contrast information, which will facilitate studies such as tumor diagnosis. FSL is a mature and widely applicable toolkit for users, with a relatively stable success rate and good performance. It has complete functions and its function-based integrated toolbox can meet the requirements of different researchers. AFNI is a flexible and complex tool that is more suitable for professional researchers. It retains most details in medical image analysis, which makes it useful in fine-grained analysis such as volume estimation. Our study provides a new idea for comparing registration tools, where tool selection strategy mainly depends on the research task in which the selected tool can leverage its unique advantages.

Identification and semi-quantification of nitrooxy organosulfates in aerosol particles by HPLC-MS/MS.

Organosulfates (OSs) derived from the oxidation of biogenic volatile organic compounds (BVOCs) in the presence of anthropogenic sulfate aerosols are the important tracers of secondary organic aerosols (SOAs). In order to better understand the concentration of pinene-nitrooxy organosulfates (pNOSs) in Nanjing, a sensitive high-performance liquid chromatography-electron spray ionization spectrum/mass spectrum (HPLC-ESI-MS/MS) to determine pNOSs in PM2.5 has been developed and validated in this study. Firstly, Hypersil Gold C18 (Thermo Scientific, San Jose, USA) was selected to separate pinene-derived nitrooxy organosulfates (pNOSs) based on their polarity. Three kinds of pNOSs were detected in the full scan mode (MS) with an ESI source under the negative mode. Secondly, three isomers of pNOSs with fragment ions m/z 220, 151, and 142 were identified based on the MS/MS maps. At least two pairs of transfer ions should be selected as identification and quantification ions according to the optimization results of target compounds. For example, to determine pNOSs, these transfer ions of m/z 294 → 247, m/z 294 → 231, m/z 294 → 220, m/z 294 → 142, m/z 294 → 151, m/z 294 → 96, m/z 294 → 80 were selected as quantification and identification ions. Finally, the influence of scan mode on pNOS detection was evaluated, and the results showed that pNOSs were most sensitive in the SRM (selected reaction monitor) scan mode. Therefore, the SRM scan mode was chosen to detect pNOSs. We applied this method to analyze year-round PM2.5 (PM2.5 is fine particulate matter, which refers to particulate matter in ambient air with an aerodynamic equivalent diameter of less than or equal to 2.5 microns) samples in Nanjing. The average concentration of all the three kinds of pNOSs was 69.95 ng m-3. The results showed that the average concentration of pNOSs was high in spring (92.94 ng m-3) and summer (90.57 ng m-3), and lowest in winter (30.03 ng m-3).

Determining Adsorption Parameters of Potentially Contaminant-Releasing Materials Using Batch Tests with Differing Liquid-Solid Ratios.

Adsorption parameters such as the distribution coefficient are required to predict the release behavior of contaminants using advection-dispersion models. However, for potentially contaminant-releasing materials (PCMs) such as dredged sludge and coal ash, these parameters cannot be obtained by conventional adsorption tests. This study developed a method to determine adsorption parameters for PCMs from a set of batch tests conducted in parallel as a function of the liquid-solid ratio (LS-parallel test). This LS-parallel test was performed on sandy soil derived from marine sediment using liquid-solid ratios from 1 to 300 L/kg. The water-contact time was also changed from 10 min to 28 d to elucidate the kinetics or equilibrium of contaminants released from the sample. Adsorption parameters were successfully obtained if the substance was under adsorption control. A column percolation test was performed to confirm the effectiveness of the obtained parameters. Good agreements were observed for SO42- and B, but discrepancies remained for other substances such as F- and As suggesting that improvements are necessary in both the LS-parallel test procedure and the advection-dispersion model.

Preparation, Bioactivities and Applications in Food Industry of Chitosan-Based Maillard Products: A Review.

Chitosan, a biopolymer possessing numerous interesting bioactivities and excellent technological properties, has received great attention from scientists in different fields including the food industry, pharmacy, medicine, and environmental fields. A series of recent studies have reported exciting results about improvement of the properties of chitosan using the Maillard reaction. However, there is a lack of a systemic review about the preparation, bioactivities and applications in food industry of chitosan-based Maillard reaction products (CMRPs). The presence of free amino groups in chitosan allows it to acquire some stronger or new functional properties via the Maillard reaction. The present review aims to focus on the current research status of synthesis, optimization and structural identification of CMRPs. The applications of CMRPs in the food industry are also discussed according to their biological and technological properties such as antioxidant, antimicrobial activities and inducing conformational changes of allergens in food. Some promising directions for future research are proposed in this review, aiming to provide theoretical guidance for the further development of chitosan and its derivatives.

Effects of bacterial-feeding nematodes and organic matter on microbial activity and oil degradation in contaminated soil.

Increasing rates of oil exploitation and utilization are associated with increasing rates of oil pollution in soil. Nematodes are abundant in soils with or without oil contamination, among which bacterial-feeding nematodes are the dominant group. However, their function in oil-contaminated soil is unclear. This study explores the effects of bacterial-feeding nematode and organic matter addition on microbial activity and oil degradation in contaminated soil. Experiments were conducted using six treatments of oil-contaminated soil: sterilized (Control), nematode-free (OC), nematode addition (OCN), nematode + wheat straw addition (OCNW), nematode + rapeseed cake addition (OCNR), and nematode + biochar addition (OCNB). At the end of a 168-day incubation experiment, the oil concentration of OCN soil was 26.77% lower than that of OC soil, and those of OCNW, OCNR, and OCNB were 12.83%, 27.81%, and 4.77% lower, respectively, than that of OCN soil. Over the experiment, soil microbial biomass carbon, fluorescein diacetate hydrolysis activity, and dehydrogenase activity increased by 4.35-382.30%, 1.75-302.22%, and - 2.73-224.55%, respectively, in oil-contaminated soils, with or without nematode and organic matter addition. These results suggest that the addition of organic matter and bacterial-feeding nematodes to oil-contaminated soil can promote the growth and activity of microorganisms that break down oil.

[Alkaline-adapted beta-mannanase of Bacillus pumilus: gene heterologous expression and enzyme characterization].

OBJECTIVE: We expressed a novel alkaline-adapted beta-mannanase gene and characterized the enzyme for potential industrial applications. METHODS: We obtained a mannanase gene (named man(B)) from Bacillus pumilus Nsic2 and expressed the gene man(B) in Escherichia coli and Bacillus subtilis. Furthermore, we characterized the enzyme. RESULTS: The gene man(B) had an open reading frame of 1104 bp that encoded a polypeptide of 367-amino-acid beta-mannanase (Man(B)). The protein sequence showed the highest identity with the beta-mannanase from B. pumilus CCAM080065. We expressed the gene man(B) in E. coli BL21 (DE3) with the enzyme activity of 11021.3 U/mL. Compared with other mannanases, Man(B) showed higher stability under alkaline conditions and was stable at pH6.0 -9.0. The specific activity of purified Man(B) was 4191 ± 107 U/mg. The K(m) and V(max) values of purified Man(B) were 35.7 mg/mL and 14.9 µmol/(mL x min), respectively. Meanwhile, we achieved recombinant protein secretion expression in B. subtilis WB800N. CONCLUSION: We achieved heterologous expression of the gene man(B) and characterized its enzyme. The alkaline-adapted Man(B) showed potential value in industrial applications due to its pH stability.

Construction of a shuttle vector for protein secretory expression in Bacillus subtilis and the application of the mannanase functional heterologous expression.

We report the construction of two Bacillus subtilis expression vectors, pBNS1/pBNS2. Both vectors are based on the strong promoter P43 and the ampicillin resistance gene expression cassette. Additionally, a fragment with the Shine-Dalgarno sequence and a multiple cloning site (BamHI, SalI, SacI, XhoI, PstI, SphI) were inserted. The coding region for the amyQ (encoding an amylase) signal peptide was fused to the promoter P43 of pBNS1 to construct the secreted expression vector pBNS2. The applicability of vectors was tested by first generating the expression vectors pBNS1-GFP/pBNS2-GFP and then detecting for green fluorescent protein gene expression. Next, the mannanase gene from B. pumilus Nsic-2 was fused to vector pBNS2 and we measured the mannanase activity in the supernatant. The mannanase total enzyme activity was 8.65 U/ml, which was 6 times higher than that of the parent strain. Our work provides a feasible way to achieve an effective transformation system for gene expression in B. subtilis and is the first report to achieve B. pumilus mannanase secretory expression in B. subtilis.