Search

We propose a novel end-to-end approach, namely, the semantic-containing double-level embedding Bi-LSTM model (SCDE-Bi-LSTM), to solve the three key problems of Q&A matching in the Chinese medical field. In the similarity calculation of the Q&A core module, we propose a text similarity calculation method that contains semantic information, to solve the problem that previous Q&A methods do not incorporate the deep information of a sentence into the similarity calculations. For the sentence vector representation module, we present a double-level embedding sentence representation method to reduce the error caused by Chinese medical word segmentation. In addition, due to the problem of the attention mechanism tending to cause backward deviation of the features, we propose an improved algorithm based on Bi-LSTM in the feature extraction stage. The Q&A framework proposed in this paper not only retains important timing features but also loses low-frequency features and noise. Additionally, it is applicable to different domains. To verify the framework, extensive Chinese medical Q&A corpora are created. We run several state-of-the-art Q&A methods as contrastive experiments on the medical corpora and the current popular insuranceQA dataset under different performance measures. The experimental results on the medical corpora show that our framework significantly outperforms several strong baselines and achieves an improvement of top-1 accuracy of up to 14%, reaching 79.15%. [ABSTRACT FROM AUTHOR]

Gestational diabetes mellitus (GDM), a high-risk pregnancy complication of great effect on the perinatal health of women and newborns, may cause changes of gut microbiota in mothers and further affect gut microbiota in newborns. This study aimed to investigate the potential effect of mother GDM on newborns’ gut microbiota. Meconium DNA was extracted from a total of 34 full-term and C-sectioned newborns, in which 20 newborns had mothers diagnosed with GDM, while 14 had unaffected mothers. Sequencing and bioinformatics analysis of 16S rRNA indicated that the gut microbiota of GDM newborns showed differences compared to control newborns. The taxonomy analyses suggested that the overall bacterial content significantly differed by maternal diabetes status, with the microbiome of the GDM group showing lower alpha-diversity than that of control group. The phyla of Proteobacteria and Actinobacteria in GDM newborns increased, while that of Bacteroidetes significantly reduced (P<0.05). Moreover, several unique gut microbiota in phylum of Proteobacteria, Firmicutes, Actinobacteria, Bacteroidetes, Chloroflexi, Acidobacteria, and Planctomycetes found in control newborns were absent in GDM ones. At genus level, the relative abundance of Prevotella and Lactobacillus significantly decreased (P<0.05) in GDM newborns. Correlation analysis indicated that maternal fasting glucose levels were positively correlated with the relative abundance of phylum Actinobacteria and genus Acinetobacter, while negatively correlated with that of phylum Bacteroidetes and genus Prevotella. However, bacteria in GDM grade A2 (GDM_A2) newborns did not show any statistical variation compared to those from control newborns, which might be attributed to the additional intervention by insulin. The results of this study have important implications for understanding the potential effects of GDM on the gut microbiota of newborns and thus possibly their metabolism at later stages in their lives. [ABSTRACT FROM AUTHOR]

We propose an improved model based on LVW embedded model feature extractor and ensemble learning for improving prediction accuracy of hemodialysis timing in this paper. Due to this drawback caused by feature extraction models, we adopt an enhanced LVW embedded model to search the feature subset by stochastic strategy, which can find the best feature combination that are most beneficial to learner performance. In the model application, we present an improved integrated learners for model fusion to reduce errors caused by overfitting problem of the single classifier. We run several state-of-the-art Q&A methods as contrastive experiments. The experimental results show that the ensemble learning model based on LVW has better generalization ability (97.04%) and lower standard error (± 0.04). We adopt the model to make high-precision predictions of hemodialysis timing, and the experimental results have shown that our framework significantly outperforms several strong baselines. Our model provides strong clinical decision support for physician diagnosis and has important clinical implications. [ABSTRACT FROM AUTHOR]

Purpose In the present study, we aimed to investigate patient-derived epilepsy-related concerns among Chinese individuals with epilepsy and the impact of seizure control on patient concerns. Methods All adult patients with epilepsy who had visited the outpatient clinic at West China Hospital between July 2014 and June 2016 were invited to participate in the study. All patients were given a blank sheet of paper and asked to list any concerns they had regarding their disease in order of importance. Demographic and clinical characteristics were also evaluated while regression models were used to examine the impact of seizure control on patient concerns. Results A total of 1040 patients reported 2202 concerns across 25 distinct categories. The most frequently listed concerns were worries about seizures (55.4%), maintaining a job (17.4%), and the heritability of epilepsy (16.0%). The legal right to drive was listed by only 3.5% of patients who took the survey while seizure recurrence was described by participants as the most important concern (379, 36.4%). Compared with the group with uncontrolled seizures, the group with controlled seizures reported less frequently about “holding down a job” (odds ratio (OR): 0.333 (0.187–0.591)), “fear of being injured during a seizure” (OR: 0.353 (0.183–0.682)), and “leading a normal life” (OR: 0.452 (0.234–0.871)), but they reported more frequently about “having another seizure” (OR: 2.447 (1.614–3.710)), “problems with medication side effects” (OR: 1.733 (1.148–2.616)), and their “legal right or ability to drive” (OR: 2.360 (1.094–5.092)). Conclusion Our findings indicated that Chinese adults with epilepsy had various concerns, some of which differed from those observed in Western populations. Concerns about heritability of seizures, marriage, and pregnancy were of greater concern to Chinese patients compared with Western patients while the legal right to drive appeared to be less of a concern to Chinese patients. Patients with controlled seizures may still have many concerns. Chinese physicians should monitor patient concerns even among those whose seizures remain controlled to meet their needs. More time and attention should be given to address these issues in clinical practice in the context of Chinese culture. [ABSTRACT FROM AUTHOR]

The aim of this study was to better understand social support in adult people with epilepsy (PWE) in China and to explore the factors related to weaker or stronger social support in PWE when compared with a group of matching healthy controls. Consecutively, we recruited PWE from the epilepsy outpatient clinic of the West China Hospital and healthy controls from nearby urban and rural areas. People with epilepsy and healthy controls were gender- and age-matched. Each participant was interviewed and completed the following instruments: the Social Support Rating Scale (SSRS) and the Hospital Anxiety and Depression Scale (HADS). In addition, we measured quality of life (QoL) in PWE using the Quality of Life in Epilepsy Inventory (QOLIE-31). We compared the SSRS scores between PWE and healthy controls and searched for relevant factors using correlation and regression analyses. The results showed that PWE scored lower on the SSRS than healthy controls. For PWE, early onset and depression were related to weaker social support. In healthy controls, being married and being psychiatrically healthy (i.e., scored lower on the HADS) were related to stronger support. Family members, especially parents and spouses, were the most powerful supporters for PWE and healthy people, but PWE relied on their families to a greater extent. Early intervention and psychiatric treatment are important to address and improve social support for PWE. [ABSTRACT FROM AUTHOR]