Modality-Aware Discriminative Fusion Network for Integrated Analysis of Brain Imaging Genomics.

Mild cognitive impairment (MCI) represents an early stage of Alzheimer's disease (AD), characterized by subtle clinical symptoms that pose challenges for accurate diagnosis. The quest for the identification of MCI individuals has highlighted the importance of comprehending the underlying mechanisms of disease causation. Integrated analysis of brain imaging and genomics offers a promising avenue for predicting MCI risk before clinical symptom onset. However, most existing methods face challenges in: 1) mining the brain network-specific topological structure and addressing the single nucleotide polymorphisms (SNPs)-related noise contamination and 2) extracting the discriminative properties of brain imaging genomics, resulting in limited accuracy for MCI diagnosis. To this end, a modality-aware discriminative fusion network (MA-DFN) is proposed to integrate the complementary information from brain imaging genomics to diagnose MCI. Specifically, we first design two modality-specific feature extraction modules: the graph convolutional network with edge-augmented self-attention module (GCN-EASA) and the deep adversarial denoising autoencoder module (DAD-AE), to capture the topological structure of brain networks and the intrinsic distribution of SNPs. Subsequently, a discriminative-enhanced fusion network with correlation regularization module (DFN-CorrReg) is employed to enhance inter-modal consistency and between-class discrimination in brain imaging and genomics. Compared to other state-of-the-art approaches, MA-DFN not only exhibits superior performance in stratifying cognitive normal (CN) and MCI individuals but also identifies disease-related brain regions and risk SNPs locus, which hold potential as putative biomarkers for MCI diagnosis.

Two novel TMEM67 variations in a Chinese family with recurrent pregnancy loss: a case report.

BACKGROUND: Recurrent pregnancy loss (RPL) is a common pregnancy complication that brings great pain to pregnant women and their families. Genetic factors are an important cause reason of RPL. However, clinical research on monogenic diseases with recurrent miscarriage is insufficient. CASE PRESENTATION: Here we reported a Chinese family with RPL and genetic analysis of the abortion and parents. A paternally inherited heterozygous missense variant c.1415T > G (p.V472G) and a maternally inherited heterozygous nonsense variant c.2314del (p.M772*) in TMEM67 gene were identified by trio-exome sequencing. c.2314del (p.M772*) generated a premature stop codon and truncated protein, was classified as "pathogenic". c.1415T > G (p.V472G) located in extra-cellular region, was classified as "likely pathogenic". Biallelic variants in TMEM67 gene cause lethal Meckel syndrome 3, consistent with the proband's prenatal phenotype. CONCLUSION: The current study of the Chinese family expands the pathogenic variant spectrum of TMEM67 and emphasizes the necessity of exome sequencing in RPL condition.

Brain Network Classification for Accurate Detection of Alzheimer's Disease via Manifold Harmonic Discriminant Analysis.

Mounting evidence shows that Alzheimer's disease (AD) manifests the dysfunction of the brain network much earlier before the onset of clinical symptoms, making its early diagnosis possible. Current brain network analyses treat high-dimensional network data as a regular matrix or vector, which destroys the essential network topology, thereby seriously affecting diagnosis accuracy. In this context, harmonic waves provide a solid theoretical background for exploring brain network topology. However, the harmonic waves are originally intended to discover neurological disease propagation patterns in the brain, which makes it difficult to accommodate brain disease diagnosis with high heterogeneity. To address this challenge, this article proposes a network manifold harmonic discriminant analysis (MHDA) method for accurately detecting AD. Each brain network is regarded as an instance drawn on a Stiefel manifold. Every instance is represented by a set of orthonormal eigenvectors (i.e., harmonic waves) derived from its Laplacian matrix, which fully respects the topological structure of the brain network. An MHDA method within the Stiefel space is proposed to identify the group-dependent common harmonic waves, which can be used as group-specific references for downstream analyses. Extensive experiments are conducted to demonstrate the effectiveness of the proposed method in stratifying cognitively normal (CN) controls, mild cognitive impairment (MCI), and AD.

GA-Net: Ghost convolution adaptive fusion skin lesion segmentation network.

Automatic segmentation of skin lesions is a pivotal task in computer-aided diagnosis, playing a crucial role in the early detection and treatment of skin cancer. Despite the existence of numerous deep learning-based segmentation methods, the extraction of lesion features remains inadequate as a result of the segmentation process. Consequently, skin lesion image segmentation continues to face challenges regarding missing detailed information and inaccurate segmentation of the lesion region. In this paper, we propose a ghost convolution adaptive fusion network for skin lesion segmentation. First, the neural network incorporates a ghost module instead of the ordinary convolution layer, generating a rich skin lesion feature map for comprehensive target feature extraction. Subsequently, the network employs an adaptive fusion module and bilateral attention module to connect the encoding and decoding layers, facilitating the integration of shallow and deep network information. Moreover, multi-level output patterns are used for pixel prediction. Layer feature fusion effectively combines output features of different scales, thus improving image segmentation accuracy. The proposed network was extensively evaluated on three publicly available datasets: ISIC2016, ISIC2017, and ISIC2018. The experimental results demonstrated accuracies of 96.42%, 94.07%, and 95.03%, and kappa coefficients of 90.41%, 81.08%, and 86.96%, respectively. The overall performance of our network surpassed that of existing networks. Simulation experiments further revealed that the ghost convolution adaptive fusion network exhibited superior segmentation results for skin lesion images, offering new possibilities for the diagnosis of skin diseases.

[Colorectal polyp segmentation method based on fusion of transformer and cross-level phase awareness].

In order to address the issues of spatial induction bias and lack of effective representation of global contextual information in colon polyp image segmentation, which lead to the loss of edge details and mis-segmentation of lesion areas, a colon polyp segmentation method that combines Transformer and cross-level phase-awareness is proposed. The method started from the perspective of global feature transformation, and used a hierarchical Transformer encoder to extract semantic information and spatial details of lesion areas layer by layer. Secondly, a phase-aware fusion module (PAFM) was designed to capture cross-level interaction information and effectively aggregate multi-scale contextual information. Thirdly, a position oriented functional module (POF) was designed to effectively integrate global and local feature information, fill in semantic gaps, and suppress background noise. Fourthly, a residual axis reverse attention module (RA-IA) was used to improve the network's ability to recognize edge pixels. The proposed method was experimentally tested on public datasets CVC-ClinicDB, Kvasir, CVC-ColonDB, and EITS, with Dice similarity coefficients of 94.04%, 92.04%, 80.78%, and 76.80%, respectively, and mean intersection over union of 89.31%, 86.81%, 73.55%, and 69.10%, respectively. The simulation experimental results show that the proposed method can effectively segment colon polyp images, providing a new window for the diagnosis of colon polyps.

Deep Manifold Harmonic Network With Dual Attention for Brain Disorder Classification.

Numerous studies have shown that accurate analysis of neurological disorders contributes to the early diagnosis of brain disorders and provides a window to diagnose psychiatric disorders due to brain atrophy. The emergence of geometric deep learning approaches provides a new way to characterize geometric variations on brain networks. However, brain network data suffer from high heterogeneity and noise. Consequently, geometric deep learning methods struggle to identify discriminative and clinically meaningful representations from complex brain networks, resulting in poor diagnostic accuracy. Hence, the primary challenge in the diagnosis of brain diseases is to enhance the identification of discriminative features. To this end, this paper presents a dual-attention deep manifold harmonic discrimination (DA-DMHD) method for early diagnosis of neurodegenerative diseases. Here, a low-dimensional manifold projection is first learned to comprehensively exploit the geometric features of the brain network. Further, attention blocks with discrimination are proposed to learn a representation, which facilitates learning of group-dependent discriminant matrices to guide downstream analysis of group-specific references. Our proposed DA-DMHD model is evaluated on two independent datasets, ADNI and ADHD-200. Experimental results demonstrate that the model can tackle the hard-to-capture challenge of heterogeneous brain network topological differences and obtain excellent classifying performance in both accuracy and robustness compared with several existing state-of-the-art methods.

Maternal pre-pregnancy/early-pregnancy smoking and risk of congenital heart diseases in offspring: A prospective cohort study in Central China.

Background: Prior studies suggested that maternal smoking before and during pregnancy could be associated with increased risks of congenital heart diseases (CHDs) in offspring. However, the results were inconsistent, and the existence of a causal relationship was not confirmed. Our study aimed to estimate the associations of maternal active and passive smoking during the pre-pregnancy/early-pregnancy period with CHDs as well as its common phenotypes in offspring. Methods: This study was based on data from a prospective cohort study conducted in Central China. A total of 49 158 eligible pregnant women between the 8th and 14th weeks of gestation were invited to join the cohort and were planned to be followed up until 3 months postpartum. The exposure of interest was maternal smoking status, including active and passive smoking status in 3 months before pregnancy as well as in early pregnancy. Self-reported maternal smoking status was ascertained via an in-person interview after recruitment. CHDs were diagnosed by pediatric cardiologists and classified according to ICD-10. Multivariable Poisson regression models were used to estimate the relative risks (RRs) with 95% confidence intervals (CIs) of all CHDs and their common phenotypes associated with maternal smoking status, adjusting for potential confounding factors identified by directed acyclic graphs. Results: CHDs were diagnosed in 564 children. After adjusting for potential confounding factors and comparing with the unexposed groups, CHDs incidence was 165% higher (adjusted RR = 2.65; 95% CI = 1.76-3.98) in offspring exposed to maternal active smoking in 3 months before pregnancy, 69% higher (adjusted-RR = 1.69; 95% CI = 1.39-2.05) in offspring exposed to maternal passive smoking in 3 months before pregnancy, 133% higher (adjusted RR = 2.33; 95% CI = 1.46-3.70) for offspring exposed to maternal active smoking in early pregnancy, and 98% higher (adjusted-RR = 1.98; 95% CI = 1.56-2.51) for offspring exposed to maternal passive smoking in early pregnancy. More specifically, the offspring exposed to maternal active smoking in early pregnancy had the highest risk of Tetralogy of Fallot (adjusted RR = 9.84; 95% CI = 2.49-38.84). These findings were recapitulated in analyses that further adjusted for other behaviour variables apart from the characteristic being assessed and were also confirmed by sensitivity analyses. Conclusions: Our findings add to the existing body of evidence that implicates maternal pre-pregnancy/early-pregnancy smoking as a significant risk factor for CHDs and their select phenotypes.

U-shaped Retinal Vessel Segmentation Based on Adaptive Aggregation of Feature Information.

Detection and analysis of retinal blood vessels contribute to the clinical diagnosis of many ophthalmic diseases. In this paper, aiming on achieving more accurate segmentation of retinal vessels and enhance the ability of the algorithm to identify microvessels, we propose a U-shaped network based on adaptive aggregation of feature information. The introduced feature selection module, which could strengthen feature transmission and selectively emphasize feature information. To effectively capture the characteristics of vessels at different scales, generate richer and denser context information, and DenseASPP is embedded at the bottom of the network. Meanwhile, we propose an adaptive aggregation module to aggregate the semantic information in each layer of the encoder part and transmit it to subsequent layers, which is beneficial to the spatial reconstruction of retinal vessels. A joint loss function is also introduced to facilitate network training. The proposed network is evaluated on three public datasets. The sensitivity, accuracy, and area under curve(AUC) are 83.48%/83.16/85.86, 95.67%/96.67%/96.52%, and 98.11%/98.69%/98.60% on DRIVE, STARE and CHASE_DB1, respectively. In order to achieve more accurate retinal blood vessel segmentation and improve the ability of the algorithm to identify microvessels. We propose a U-shaped network based on adaptive aggregation of feature information. The introduction of the adaptive aggregation module aggregates the semantic information of each level of the encoder part, which improves the robustness of the model to segment blood vessels.

Association Between First-Trimester Maternal Cytomegalovirus Infection and Stillbirth: A Prospective Cohort Study.

Background: Given that the time lag between cytomegalovirus (CMV) screening and diagnosed testing, a better knowledge of the association between pregnant women with CMV screening test positive and stillbirth in an epidemiological perspective was required to assist people being counseled reframe their pregnancy and birth plans based on the magnitude of the risk. Methods: This study recruited 44048 eligible pregnant women from March 13, 2013 to December 31, 2019. Serological tests including CMV-specific IgM and IgG, and IgG avidity index were used to screen for maternal CMV infection and were measured by automated chemiluminescence immunoassay. The association was assessed using the inverse probability of group-weighted multivariate-adjusted log-binomial models. Results: A total of 540 infants ended with a stillbirth (12.3 per 1000 pregnancies), and 2472 pregnancies with maternal CMV infection were screened out (56.1 per 1000 pregnancies) among all eligible pregnancies. In the comparison analysis, 326 infants ended with a stillbirth (86.6 per 1000 pregnancies) in the maternal CMV infection group compared with 214 infants (7.8 per 1000 pregnancies) in the group where mothers were not infected with CMV (RR 12.17; 95% CI 9.43-15.71). After excluding the pregnancies of stillbirth with birth defects, a strong association between the two groups was still observed (RR 9.38; 95% CI 6.92-12.70). Conclusion: Our findings quantified the risk of a woman having a baby with stillbirth if she had a positive serologic CMV screening test in her first trimester, and supported the value of using CMV serologic tests as part of regular testing in pregnant women. Trial registration: Registered in Chinese Clinical Trial Registry Center; registration number, ChiCTR1800016635; registration date, 06/14/2018 (Retrospectively registered); URL of trial registry record, https://www.chictr.org.cn/showproj.aspx?proj=28300.

Maternal Viral Infection in Early Pregnancy and Risk of Congenital Heart Disease in Offspring: A Prospective Cohort Study in Central China.

PURPOSE: To examine the associations of maternal virus infection in early pregnancy with risk of offspring congenital heart disease (CHD) and its seven common subtypes including atrial septal defect, ventricular septal defect, atrioventricular septal defect, patent ductus arteriosus, Tetralogy of Tallot, pulmonary stenosis, and transposition of the great arteries. PATIENTS AND METHODS: A prospective cohort study was conducted in Central China. A total of 44,048 pregnant women with singleton pregnancies at 8-14 gestational weeks were finally included and followed to 3 months postpartum. Serum was tested for virus infection including hepatitis B virus (HBV), coxsackievirus-B, human cytomegalovirus (HCMV), herpes simplex virus (HSV), and rubella virus. Multivariable modified Poisson regression models were used to estimate the relative risks (RRs) of all CHDs as well as seven common subtypes of CHD in offspring of pregnant women with different types of virus infection in early pregnancy, adjusting for confounders identified by directed acyclic graphs. RESULTS: At the end of follow-up, 564 births were diagnosed with CHD. Multivariable analyses showed that the presence of maternal viral infection in early pregnancy was independently associated with increased risks of CHD in offspring, with an adjusted relative risk of 2.21 (95% CI: 1.66-2.95) for HBV infection, 2.21 (95% CI: 1.63-3.00) for coxsackievirus-B infection, 3.12 (95% CI: 2.44-3.98) for HCMV infection, and 2.62 (95% CI: 1.95-3.51) for rubella virus infection. More specifically, the offspring of pregnant women with HCMV infection had the highest increased risk of patent ductus arteriosus (RR=10.50, 95% CI: 6.24-17.66). These findings persisted in analyses that were further adjusted for the other virus of interest in this study. CONCLUSION: Our study proposed evidence that maternal virus infection in early pregnancy, including HBV, coxsackievirus-B, HCMV, and rubella virus, was implicated in CHD, although more studies remain needed to verify the results, especially associations in specific CHD phenotypes.

Effect of maternal alcohol consumption during the pre-pregnancy/early-pregnancy period on congenital heart disease: A prospective cohort study in Central China.

Evidence of associations between maternal alcohol consumption and congenital heart disease (CHD) are mixed. Previous studies have been potentially biased due to recall bias or unmeasured confounding. This study aimed to examine the association of maternal alcohol consumption in 3 months before pregnancy and in early pregnancy with risks of offspring congenital heart disease (CHD) and its seven common subtypes. A prospective cohort study was conducted in Central China. From 03/13/2013 to 12/31/2019, a total of 44,048 pregnant women with singleton pregnancies at 8-14 gestational weeks were included and followed to 3 months postpartum. 564 births were diagnosed with CHD at the end of follow-up. Multivariable modified Poisson regression models were used to estimate the relative risks (RRs) of CHD in offspring exposed to maternal alcohol consumption during the pre-pregnancy and early-pregnancy period, adjusting for confounders identified by directed acyclic graphs. In the multivariable analyses, increased risks of CHDs were found in offspring exposed to maternal alcohol consumption both in 3 months before pregnancy (adjusted-RR:3.14; 95% confidence intervals[CIs]:2.30-4.28) and in early pregnancy (adjusted-RR:1.86; 95%CIs:1.13-3.05). More specifically, the offspring exposed to maternal alcohol consumption in 3 months before pregnancy had the highest increased risk of Tetralogy of Fallot (adjusted-RR:8.62; 95%CIs:3.61-20.61). These findings persisted in analyses that were further adjusted for the other behavior variables other than the characteristic being assessed, and were also confirmed by sensitivity analyses. Our study supports the need for continued efforts for public health messages surrounding the potential risks of alcohol consumption prior to or during pregnancy.

Deviated and early unsustainable stunted development of gut microbiota in children with autism spectrum disorder.

OBJECTIVE: Recent studies have provided insights into the gut microbiota in autism spectrum disorder (ASD); however, these studies were restricted owing to limited sampling at the unitary stage of childhood. Herein, we aimed to reveal developmental characteristics of gut microbiota in a large cohort of subjects with ASD combined with interindividual factors impacting gut microbiota. DESIGN: A large cohort of 773 subjects with ASD (aged 16 months to 19 years), 429 neurotypical (NT) development subjects (aged 11 months to 15 years) were emolyed to determine the dynamics change of gut microbiota across different ages using 16S rRNA sequencing. RESULT: In subjects with ASD, we observed a distinct but progressive deviation in the development of gut microbiota characterised by persistently decreased alpha diversity, early unsustainable immature microbiota, altered aboudance of 20 operational taxonomic units (OTUs), decreased taxon detection rate and 325 deregulated microbial metabolic functions with age-dependent patterns. We further revealed microbial relationships that have changed extensively in ASD before 3 years of age, which were associated with the severity of behaviour, sleep and GI symptoms in the ASD group. This analysis demonstrated that a signature of the combination of 2 OTUs, Veillonella and Enterobacteriaceae, and 17 microbial metabolic functions efficiently discriminated ASD from NT subjects in both the discovery (area under the curve (AUC)=0.86), and validation 1 (AUC=0.78), 2 (AUC=0.82) and 3 (AUC=0.67) sets. CONCLUSION: Our large cohort combined with clinical symptom analysis highlights the key regulator of gut microbiota in the pathogenesis of ASD and emphasises the importance of monitoring and targeting the gut microbiome in future clinical applications of ASD.

Accidental death during pregnancy and puerperium from 2009 to 2019 in Hunan, China: a cross-sectional study.

OBJECTIVE: Few studies have analysed accidental maternal deaths. This study analysed the basic situation and classification of maternal accidental deaths and compared the differences between urban and rural areas. DESIGN: A cross-sectional study on accidental deaths during pregnancy and puerperium from 2009 to 2019 in Hunan Province. SETTING: Hunan Province, with a population of 74 million, has an area of 210 000 km2 and 123 counties/districts. PARTICIPANTS: A collection of 239 cases of accidental death during pregnancy and puerperium in Hunan Province from 2009 to 2019, including 181 cases of rural pregnancy and puerperium and 58 cases of urban pregnancy and puerperium. MAIN OUTCOME MEASURE: Classification of accidental mortality of pregnant women. RESULTS: A total of 239 accidental deaths occurred in Hunan Province, with an accidental mortality rate of 2.8 per 100 000 live births. The accidental mortality rate in rural areas (3.2 per 100 000 live births) was higher than in urban areas (2.0 per 100 000 live births). The proportion of accidental deaths among pregnancy-related deaths showed an upward trend. The main types of accidental deaths were suicide (1.0 per 100 000 live births), traffic accidents (0.8 per 100 000 live births), accidental poisoning/overdose and assault/homicide (0.2 per 100 000 live births), and other accidents (0.6 per 100 000 live births). Maternal accidental deaths were mainly concentrated in low-income families, in rural areas and in those with low level of education. 74.5% of accidental deaths occurred before childbirth. 49.2% of pregnant women gave birth by caesarean section. CONCLUSION: In response to the different causes of accidental maternal death, public health programmes and policy interventions should pay special attention to maternal suicide and traffic injuries.

[Association between maternal reduced folate carrier gene polymorphisms and congenital heart disease in offspring: a case-control study].

OBJECTIVE: To study the association between maternal reduced folate carrier (RFC) gene polymorphisms and congenital heart disease (CHD) in offspring. METHODS: A hospital-based case-control study was conducted. The mothers of 683 infants with CHD who attended the Department of Cardiothoracic Surgery, Hunan Children's Hospital, from November 2017 to March 2020 were enrolled as the case group. The mothers of 740 healthy infants without any deformity who attended the hospital during the same period of time were enrolled as the control group. A questionnaire survey was performed to collect the exposure data of subjects. Venous blood samples of 5 mL were collected from the mothers for genetic polymorphism detection. A multivariate logistic regression analysis was used to evaluate the association of RFC gene polymorphisms and their haplotypes with CHD. A generalized multifactor dimensionality reduction method was used to analyze gene-gene interactions. RESULTS: After control for confounding factors, the multivariate logistic regression analysis showed that maternal RFC gene polymorphisms at rs2236484 (AG vs AA:OR=1.91, 95%CI:1.45-2.51; GG vs AA: OR=1.96, 95%CI:1.40-2.75) and rs2330183 (CT vs CC:OR=1.39, 95%CI:1.06-1.83) were significantly associated with the risk of CHD in offspring. The haplotypes of G-G (OR=1.21, 95%CI:1.03-1.41) and T-G (OR=1.25, 95%CI:1.07-1.46) in mothers significantly increased the risk of CHD in offspring. The interaction analysis showed significant gene-gene interactions between different SNPs of the RFC gene in CHD (P < 0.05). CONCLUSIONS: Maternal RFC gene polymorphisms and interactions between different SNPs are significantly associated with the risk of CHD in offspring.

Association analysis of maternal MTHFR gene polymorphisms and the occurrence of congenital heart disease in offspring.

BACKGROUND: Although many studies showed that the risk of congenital heart disease (CHD) was closely related to genetic factors, the exact pathogenesis is still unknown. Our study aimed to comprehensively assess the association of single nucleotide polymorphisms (SNPs) of maternal MTHFR gene with risk of CHD and its three subtypes in offspring. METHODS: A case-control study involving 569 mothers of CHD cases and 652 health controls was conducted. Thirteen SNPs were detected and analyzed. RESULTS: Our study showed that genetic polymorphisms of maternal MTHFR gene at rs4846052 and rs1801131 were significantly associated with risk of CHD in the homozygote comparisons (TT vs. CC at rs4846052: OR = 7.62 [95%CI 2.95-19.65]; GG vs. TT at rs1801131: OR = 5.18 [95%CI 2.77-9.71]). And six haplotypes of G-C (involving rs4846048 and rs2274976), A-C (involving rs1801133 and rs4846052), G-T (involving rs1801133 and rs4846052), G-T-G (involving rs2066470, rs3737964 and rs535107), A-C-G (involving rs2066470, rs3737964 and rs535107) and G-C-G (involving rs2066470, rs3737964 and rs535107) were identified to be significantly associated with risk of CHD. Additionally, we observed that a two-locus model involving rs2066470 and rs1801131 as well as a three-locus model involving rs227497, rs1801133 and rs1801131 were significantly associated with risk of CHD in the gene-gene interaction analyses. For three subtypes including atrial septal defect, ventricular septal defect and patent ductus arteriosus, similar results were observed. CONCLUSIONS: Our study indicated genetic polymorphisms of maternal MTHFR gene were significantly associated with risk of fetal CHD in the Chinese population. Additionally, there were significantly interactions among different SNPs on risk of CHD. However, how these SNPs affect the development of fetal heart remains unknown, and more studies in different ethnic populations and with a larger sample are required to confirm these findings.

Clinical significance of S100B protein in pregnant woman with early- onset severe preeclampsia.

OBJECTIVES: Preeclampsia is one of the most feared complications of pregnancy, which can progress rapidly to serious complications such as death of both mother and fetus. To present, the leading cause of preeclampsia is still debated. The purpose of this article was to explore the clinical significance of S100B protein, a kind of Ca2+ -sensor protein, in the early-onset severe preeclampsia. MATERIAL AND METHODS: Nine pregnant women with early-onset severe preeclampsia (the study group) and 13 healthy pregnant women (the control group) were included in this study. The level of S100B in the amniotic fluid, maternal blood, and umbilical cord blood were detected by enzyme-linked immunosorbent assay (ELISA) and surface plasmon resonance imaging (SPRi) methods. Diagnostic values of S100B for early-onset severe preeclampsia were assessed by Receiver Operating Characteristic (ROC) curve analysis. RESULTS: The levels of S100B in maternal blood and amniotic fluid in the study group were higher than those in the control group (p < 0.05). ROC curve analysis showed that S100B detected by SPRi method (SPRi-S100B) showed a cut-off level of 181 ng/mL with sensitivity of 100%, a specificity of 84.6%, and a Youden index of 0.846 in the maternal blood, which had better clinical significance and diagnostic value (at than that detected by ELISA (ELISA-S100B). CONCLUSIONS: The levels of S100B detected by SPRi in maternal blood can indicate early-onset severe preeclampsia and perinatal brain injury.

[Segmentation of retinal vessels by fusing contour information and conditional generative adversarial].

The existing retinal vessels segmentation algorithms have various problems that the end of main vessels are easy to break, and the central macula and the optic disc boundary are likely to be mistakenly segmented. To solve the above problems, a novel retinal vessels segmentation algorithm is proposed in this paper. The algorithm merged together vessels contour information and conditional generative adversarial nets. Firstly, non-uniform light removal and principal component analysis were used to process the fundus images. Therefore, it enhanced the contrast between the blood vessels and the background, and obtained the single-scale gray images with rich feature information. Secondly, the dense blocks integrated with the deep separable convolution with offset and squeeze-and-exception (SE) block were applied to the encoder and decoder to alleviate the gradient disappearance or explosion. Simultaneously, the network focused on the feature information of the learning target. Thirdly, the contour loss function was added to improve the identification ability of the blood vessels information and contour information of the network. Finally, experiments were carried out on the DRIVE and STARE datasets respectively. The value of area under the receiver operating characteristic reached 0.982 5 and 0.987 4, respectively, and the accuracy reached 0.967 7 and 0.975 6, respectively. Experimental results show that the algorithm can accurately distinguish contours and blood vessels, and reduce blood vessel rupture. The algorithm has certain application value in the diagnosis of clinical ophthalmic diseases.

Identification of gender-related metabolic disturbances in autism spectrum disorders using urinary metabolomics.

Autism spectrum disorders (ASD) are a highly heterogeneous group of neurodevelopmental disorders that are more commonly diagnosed in boys than in girls. The reasons for gender differences in ASD are unknown and no definitive current evidence can explain male predominance. Therefore, in search for laboratory biomarkers responsible for ASD, a comprehensive metabolomics study was performed by metabolic profiling of urine samples in 51 ASD subjects and 51 age- and sex-matched children with typical development. Orthogonal partial least-squares discriminant analysis (OPLS-DA) models with poor quality failed to perform the analysis based on gender in the ASD and control groups. OPLS-DA models based on single-sex samples, especially in female subjects, had better clustering between the ASD and control groups with an increase in the R2 and Q2 values compared with those in the whole group. Significantly increased levels of adenine, 2-Methylguanosine, creatinine, and 7alpha-hydroxytestololactone and a decrease in creatine were observed in the female ASD subjects. In particular, 7alpha-hydroxytestololactone, which has a structure similar to that of testolactone, was positively correlated with adenine (Pearson correlation coefficient, r = 0.738, p < 0.01), creatinine (r = 0.826, p < 0.01), and 2-Methylguanosine (r = 0.757, p < 0.01) and negatively correlated with creatine (r=-0.413, p < 0.05). A receiver operating characteristic curve analysis using the creatinine:creatine ratio yielded an area under the curve of 0.913 (95% CI: 0.806-1). These metabolites may be sex-related or sex-sensitive to an extent and can be valuable for identification of the molecular pathways involved in the gender bias in manifestation of ASD. The creatinine:creatine ratio has a potential to be a good predictor of ASD in the female subjects.

Epidemiological characteristics and influencing factors of fatal drowning in children under 5 years old in Hunan Province, China: case-control study.

BACKGROUND: Drowning is still the primary cause of death in children under 5 years old, however, little is known about the drowning of Hunan province children. This study identifies the previously unpublished incidence and characteristics of fatal drowning in children of Hunan Province, and provide a basis for formulating strategies for children's survival, development and protection. METHODS: Data were collected through sampling with the multistage stratified cluster. The case group included all fatal frowning children under 5 years old in 13 districts between October 2015 and September 2016. The control group was matched 1:1.The epidemic features and influencing factors of fatal drowning were analyzed retrospectively according to descriptive analysis, conditional univariate and multivariate logistic regression analysis. RESULTS: For children aged 0-4 years, the fatal drowning was 16.1/100000 in Hunan Province. Drowning rates were higher for boys than girls. The proportion of rural areas is much higher than that of urban areas. The 1-2 years age-group was the highest of all age groups. Fatal drowning mainly occurred in summer. The three leading drowning locations were pond, ditch and well. Playing close to the water were the leading activities that preceded fatal drowning. Multivariate logistic regression analysis showed that: children with primary caregiver education in high school and above (OR = 0.05) have a lower risk of fatal drowning; children with full-time care (OR = 0.17) have a lower risk; children who received unintentional drowning safety education (OR = 0.23) have a lower risk of fatal drowning. Children who were always swimming or playing near the water in the past 6 months (OR = 3.13) have a higher risk of fatal drowning. CONCLUSION: The fatal drowning among children under 5 years is the result of the interaction of multiple factors. A significant number of child deaths could have been prevented if parents and other close relatives had been more concerned about the safety of their children. We should develop health education plans for villagers to warn them of the dangers of drowning and preventive measures.

A Preliminary Study of Uric Metabolomic Alteration for Postpartum Depression Based on Liquid Chromatography Coupled to Quadrupole Time-of-Flight Mass Spectrometry.

Postpartum depression affects about 10-20% of newly delivered women, which is harmful for both mothers and infants. However, the current diagnosis of postpartum depression depends on the subjective judgment of a practitioner, which may lead to misdiagnosis. Hence, an appended objective diagnosis index may help the practitioner to improve diagnosis. A metabolomic study can find biomarkers as an objective index to facilitate disease diagnosis. Forty-nine postpartum depressed patients and 50 healthy controls were recruited into this study. The metabolites in urine were scanned with LC-Q-TOF-MS. The metabolomic data were analyzed with a multivariate statistical analysis method. Data from 40 patients and 40 controls were used for partial least square-discriminate analysis (PLS-DA). The urine metabolomic profiles of patients were different from those of controls. The PLS-DA model was validated by a permutation test, and the model could accurately classify the other 9 patients and 10 controls in T-prediction. Ten differentiating metabolites were found as main contributors to this difference, which are involved in amino acid metabolism, neurotransmitter metabolism, bacteria population, etc. Some of these potential biomarkers, such as 4-hydroxyhippuric acid, homocysteine, and tyrosine, showed relatively high sensitivities and specificities. The metabolic profile alteration induced by postpartum depression was found, and some of the differentiating metabolites may serve as biomarkers to facilitate the diagnosis of postpartum depression.