Predicting preterm birth using auto-ML frameworks: a large observational study using electronic inpatient discharge data.

Background: To develop and compare different AutoML frameworks and machine learning models to predict premature birth. Methods: The study used a large electronic medical record database to include 715,962 participants who had the principal diagnosis code of childbirth. Three Automatic Machine Learning (AutoML) were used to construct machine learning models including tree-based models, ensembled models, and deep neural networks on the training sample (N = 536,971). The area under the curve (AUC) and training times were used to assess the performance of the prediction models, and feature importance was computed via permutation-shuffling. Results: The H2O AutoML framework had the highest median AUC of 0.846, followed by AutoGluon (median AUC: 0.840) and Auto-sklearn (median AUC: 0.820), and the median training time was the lowest for H2O AutoML (0.14 min), followed by AutoGluon (0.16 min) and Auto-sklearn (4.33 min). Among different types of machine learning models, the Gradient Boosting Machines (GBM) or Extreme Gradient Boosting (XGBoost), stacked ensemble, and random forrest models had better predictive performance, with median AUC scores being 0.846, 0.846, and 0.842, respectively. Important features related to preterm birth included premature rupture of membrane (PROM), incompetent cervix, occupation, and preeclampsia. Conclusions: Our study highlights the potential of machine learning models in predicting the risk of preterm birth using readily available electronic medical record data, which have significant implications for improving prenatal care and outcomes.

Prenatal Genetic Diagnosis in Three Fetuses With Left Heart Hypoplasia (LHH) From Three Unrelated Families.

Background: Congenital heart defects (CHDs) are the most common birth defects, and left heart hypoplasia (LHH) is a severe form of CHD and responsible for more than 20% cardiac deaths during the first week of life, however, its genetic causes remain largely elusive. Methods: Three families with fetal LHH were recruited. Genomic DNA from amniotic fluid or peripheral blood, and trio whole exome sequencing (trio-WES) and copy number variation sequencing (CNV-seq) were performed. Results: All the three couples had no family history, and mid-gestation ultrasound revealed LHH and other variable cardiovascular defects in the fetuses. Trio-WES revealed de novo pathogenic variations in KMT2D (p.Gly3465Aspfs*37) (NM_003482) and WDFY3 (p.Ser117Xfs*) (NM_014991), and CNV-seq identified a deletion of 150 kb encompassing NOTCH1. KMT2D and NOTCH1 previously have been reported to be associated with CHDs, however, WDFY3 is reported for the first time to be possibly related to CHD in human. Conclusion: Our study suggested that genetic component is an important risk factor for the development of LHH, and next generation sequencing is a powerful tool for genetic diagnosis in fetuses with CHDs and genetic counseling, however, more studies and data are need to establish the correlation of fetal phenotypes and genotypes.

Knowledge, attitudes, practices, and influencing factors of anxiety among pregnant women in Wuhan during the outbreak of COVID-19: a cross-sectional study.

BACKGROUND: Prenatal anxiety has been a significant public health issue globally, leading to adverse health outcomes for mothers and children. The study aimed to evaluate the sociodemographic characteristics, knowledge, attitudes, and practices (KAP), and anxiety level of pregnant women during the coronavirus disease 2019 (COVID-19) epidemic in Wuhan and investigate the influencing factors for prenatal anxiety in this specific context. METHODS: Pregnant subjects' KAP towards COVID-19 and their sociodemographics and pregnancy information were collected using questionnaires. The Zung Self-Rating Anxiety Scale (SAS) was used to assess anxiety status. Factors associated with the level of prenatal anxiety were analyzed by Pearson's chi-square test and multivariable logistic regression analyses. RESULTS: The prenatal anxiety prevalence in this population was 20.8%. The mean score of knowledge was 13.2 ± 1.1 on a 0 ~ 14 scale. The attitudes and practices data showed that 580/ 817 (71.0%) were very concerned about the news of COVID-19, 455/817 (55.7%) considered the official media to be the most reliable information source for COVID-19, and 681/817 (83.4%) were anxious about the possibility of being infected by COVID-19. However, only 83/817 (10.2%) worried about contracting COVID-19 infection through the ultrasound transducer during a routing morphology scan. About two-thirds 528/817 (64.6%) delayed or canceled the antenatal visits. Approximately half of them 410/817 (50.2%) used two kinds of personal protection equipments (PPEs) during hospital visits. Logistic regression analysis revealed that the influential factors for prenatal anxiety included previous children in the family, knowledge score, media trust, worry of contracting the COVID-19 infection and worry about getting infected with COVID-19 from the ultrasound probe antenatal care (ANC) schedule. CONCLUSION: Prenatal anxiety was prevalent among pregnant women in Wuhan during the outbreak of COVID-19. The current findings identified factors associated with the level of prenatal anxiety that could be targeted for psychological care.

Magnetic iron oxide nanoparticles mediated gene therapy for breast cancer--an in vitro study.

The aim of this study was to evaluate the feasibility and efficacy of using TRAIL gene to treat breast cancer mediated with a novel carrier - magnetic iron oxide nanoparticles (poly-MAG-1000) coated with PEI. The magnetic iron oxide nanoparticles were used as gene carrier to transfect TRAIL gene into MCF-7 cells. The polyMAG-1000 without TRAIL gene was transfected into the tumor cells as negative control. TRAIL gene transfection with liposome as carrier served as positive control. The apoptosis of cells was detected with TUNEL method. The apoptosis ratio of tumor cells was measured with flow cytometry (FCM). It was found that the apoptosis occurred in the tumor cells after transfection of TRAIL gene mediated by both polyMAG-1000 and liposome. The apoptosis ratio in the group with polyMAG-1000 as gene carrier was (25.11+/-2.85) %, whereas it was (5.06+/- 1.05) % in the control group with polyMAG-1000 (P<0.01). The apoptosis ratio was as low as (18.31+/-2.44) % in the group with liposome as gene carrier (P<0.05, as compared with the group with polyMAG-1000 as gene carrier). It is suggested that TRAIL gene may induce apoptosis in MCF-7 breast cancer cells. The magnetic iron oxide nanoparticles coated with PEI may be a potential gene carrier with high transfection efficacy for cancer gene therapy..

Immune vasculitis induced atherosclerosis.

The relationship between immune vasculitis and atherosclerosis was studied. The experimental model of weanling rabbits for immune vasculitis was reproduced by intravenous injection of 10% bovine serum albumin. There were 6 groups: group A, 25 weanling rabbits with immune vasculitis subject to coronary arteriography; group B, 10 normal mature rabbits subject to coronary arteriography; group C, 10 weanling rabbits subject to coronary arteriography; group D, 8 weanling rabbits with vasculitis and cholesterol diet; group E, 8 weanling rabbits receiving single cholesterol diet; group F: 8 weanling rabbits receiving basic diet. Four weeks later, coronary arteriography was performed in groups A, B and C. The rabbits in groups D, E and F were sacrificed for the study of pathological changes in the coronary artery after 12 weeks. The results showed that the dilatation of coronary artery occurred in 6 rabbits of group A, but in groups B and C, no dilatation of coronary artery appeared. In comparison with group E, more severe atherosclerosis occurred in group D, showing the thickened plaque, fibrous sclerosis and atherosclerotic lesion. Percentage of plaques covering aortic intima, incidence of atherosclerosis of small coronary arteries and degree of stenosis of coronary arteries were significantly higher in group D than in group E (P < 0.01). No atherosclerosis changes were found in group F. It was concluded that in the acute phase, the serum immune vasculitis can induce the dilatation of coronary artery of some weanling rabbits, and aggravate the formation of atherosclerosis in rabbits fed with cholesterol diet. Immune vasculitis is a new risk factor of atherosclerosis and ischemic heart disease.

Use of PEI-coated magnetic iron oxide nanoparticles as gene vectors.

To evaluate the feasibility of using polyethyleneimine (PEI) coated magnetic iron.oxide nanoparticles (polyMAG-1000) as gene vectors. The surface characteristics of the nanoparticles were observed with scanning electron microscopy. The ability of the nanoparticles to combine with and protect DNA was investigated at different PH values after polyMAG-1000 and DNA were combined in different ratios. The nanoparticles were tested as gene vectors with in vitro transfection models. Under the scanning electron microscope the nanoparticles were about 100 nm in diameter. The nanoparticles could bind and condense DNA under acid, neutral and alkaline conditions, and they could transfer genes into cells and express green fluorescent proteins (GFP). The transfection efficiency was highest (51%) when the ratio of nanoparticles to DNA was 1:1 (v:w). In that ratio, the difference in transfection efficiency was marked depending on whether a magnetic field was present or not: about 10% when it was absent but 51% when it was present. The magnetic iron oxide nanoparticles coated with PEI may potentially be used as gene vectors.