Sleep staging is the basis for solving sleep problems. There's an upper limit for the classification accuracy of sleep staging models based on single-channel electroencephalogram (EEG) data and features. To address this problem, this paper proposed an automatic sleep staging model that mixes deep convolutional neural network (DCNN) and bi-directional long short-term memory network (BiLSTM). The model used DCNN to automatically learn the time-frequency domain features of EEG signals, and used BiLSTM to extract the temporal features between the data, fully exploiting the feature information contained in the data to improve the accuracy of automatic sleep staging. At the same time, noise reduction techniques and adaptive synthetic sampling were used to reduce the impact of signal noise and unbalanced data sets on model performance. In this paper, experiments were conducted using the Sleep-European Data Format Database Expanded and the Shanghai Mental Health Center Sleep Database, and achieved an overall accuracy rate of 86.9% and 88.9% respectively. When compared with the basic network model, all the experimental results outperformed the basic network, further demonstrating the validity of this paper's model, which can provide a reference for the construction of a home sleep monitoring system based on single-channel EEG signals.
China ; Sleep Stages ; Sleep ; Electroencephalography ; Databases, Factual

Arrhythmia is a significant cardiovascular disease that poses a threat to human health, and its primary diagnosis relies on electrocardiogram (ECG). Implementing computer technology to achieve automatic classification of arrhythmia can effectively avoid human error, improve diagnostic efficiency, and reduce costs. However, most automatic arrhythmia classification algorithms focus on one-dimensional temporal signals, which lack robustness. Therefore, this study proposed an arrhythmia image classification method based on Gramian angular summation field (GASF) and an improved Inception-ResNet-v2 network. Firstly, the data was preprocessed using variational mode decomposition, and data augmentation was performed using a deep convolutional generative adversarial network. Then, GASF was used to transform one-dimensional ECG signals into two-dimensional images, and an improved Inception-ResNet-v2 network was utilized to implement the five arrhythmia classifications recommended by the AAMI (N, V, S, F, and Q). The experimental results on the MIT-BIH Arrhythmia Database showed that the proposed method achieved an overall classification accuracy of 99.52% and 95.48% under the intra-patient and inter-patient paradigms, respectively. The arrhythmia classification performance of the improved Inception-ResNet-v2 network in this study outperforms other methods, providing a new approach for deep learning-based automatic arrhythmia classification.
Humans ; Arrhythmias, Cardiac/diagnostic imaging* ; Cardiovascular Diseases ; Algorithms ; Databases, Factual ; Electrocardiography

In the diagnosis of cardiovascular diseases, the analysis of electrocardiogram (ECG) signals has always played a crucial role. At present, how to effectively identify abnormal heart beats by algorithms is still a difficult task in the field of ECG signal analysis. Based on this, a classification model that automatically identifies abnormal heartbeats based on deep residual network (ResNet) and self-attention mechanism was proposed. Firstly, this paper designed an 18-layer convolutional neural network (CNN) based on the residual structure, which helped model fully extract the local features. Then, the bi-directional gated recurrent unit (BiGRU) was used to explore the temporal correlation for further obtaining the temporal features. Finally, the self-attention mechanism was built to weight important information and enhance model's ability to extract important features, which helped model achieve higher classification accuracy. In addition, in order to mitigate the interference on classification performance due to data imbalance, the study utilized multiple approaches for data augmentation. The experimental data in this study came from the arrhythmia database constructed by MIT and Beth Israel Hospital (MIT-BIH), and the final results showed that the proposed model achieved an overall accuracy of 98.33% on the original dataset and 99.12% on the optimized dataset, which demonstrated that the proposed model can achieve good performance in ECG signal classification, and possessed potential value for application to portable ECG detection devices.
Humans ; Electrocardiography ; Algorithms ; Cardiovascular Diseases ; Databases, Factual ; Neural Networks, Computer

Objective: To analyze the global epidemiology of renal cell carcinoma (RCC) in 2020. Methods: The incidence and mortality data of RCC in the cooperative database GLOBOCAN 2020 of International Agency for Research on Cancer of WHO and the human development index (HDI) published by the United Nations Development Programme in 2020 were collated. The crude incidence rate (CIR), age-standardized incidence rate (ASIR), crude mortality rate (CMR), age-standardized mortality rate (ASMR) and mortality/incidence ratio (M/I) of RCC were calculated. Kruskale-Wallis test was used to analyze the differences in ASIR or ASMR among HDI countries. Results: In 2020, the global ASIR of RCC was 4.6/100 000, of which 6.1/100 000 for males and 3.2/100 000 for females and ASIR was higher in very high and high HDI countries than that in medium and low HDI countries. With the rapid increase of age after the age of 20, the growth rate of ASIR in males was faster than that in females, and slowed down at the age of 70 to 75. The truncation incidence rate of 35-64 years old was 7.5/100 000 and the cumulative incidence risk of 0-74 years old was 0.52%. The global ASMR of RCC was 1.8/100 000, 2.5/100 000 for males and 1.2/100 000 for females. The ASMR of males in very high and high HDI countries (2.4/100 000-3.7/100 000) was about twice that of males (1.1/100 000-1.4/100 000) in medium and low HDI countries, while the ASMR of female (0.6/100 000-1.5/100 000) did not show significant difference. ASMR continued to increase rapidly with age after the age of 40, and the growth rate of males was faster than that of females. The truncation mortality rate of 35-64 years old was 2.1/100 000, and the cumulative mortality risk of 0-74 years old was 0.20%. M/I decreases with the increase of HDI, with M/I as 0.58 in China, which was higher than the global average of 0.39 and the United States' 0.17. Conclusion: The ASIR and ASMR of RCC presented significant regional and gender disparities globally, and the heaviest burden was in very high HDI countries.
Male ; Humans ; Female ; Adult ; Middle Aged ; Infant, Newborn ; Infant ; Child, Preschool ; Child ; Adolescent ; Young Adult ; Aged ; Carcinoma, Renal Cell/epidemiology* ; Incidence ; Databases, Factual ; China ; Kidney Neoplasms/epidemiology* ; Global Health

Objective: To understand the status of autism spectrum disorder (ASD) cohort studies and explore the feasibility of constructing ASD disease-specific cohorts based on real-world data (RWD). Methods: ASD cohort studies published by December 2022 were collected by literature retrieval from major Chinese and English databases. And the characteristics of the cohort were summarized. Results: A total of 1 702 ASD cohort studies were included, and only 60 (3.53%) were from China. A total of 163 ASD-related cohorts were screened, of which 55.83% were birth cohorts, 28.22% were ASD-specific cohorts, and 4.91% were ASD high-risk cohorts. Most cohorts used RWD such as hospital registries or conducted community-based field surveys to obtain participant information and identified patients with ASD by scales or clinical diagnoses. The contents of the studies included ASD incidence and prognostic risk factors, ASD comorbidity patterns and the impact of ASD on self-health and their offspring's health. Conclusions: ASD cohort studies in developed countries have been in the advanced stage, while the Chinese studies are still in their infancy. RWD provides the data basis for ASD-specific cohort construction and offers new opportunities for research, but work such as case validation is still needed to ensure the scientific nature of cohort construction.
Humans ; Autism Spectrum Disorder ; Cohort Studies ; Databases, Factual

Artificial intelligence revealed its application prospects that could bring change in oral medicine. Artificial intelligence related papers in oral medicine field increased year by year since the 1990s. In order to provide reference for further research, the literature on artificial intelligence studies and its application in oral medicine were retrieved from multiple databases and summarized. The evolution of hot spots on artificial intelligence and related state of the art technology in oral medicine were analyzed.
Artificial Intelligence ; Databases, Factual ; Oral Medicine ; Technology ; Humans

OBJECTIVE: To investigate the prognostic value of death-associated protein 5 (DAP5) in gastric cancer (GC) and its regulatory effect on aerobic glycolysis in GC cells. METHODS: We analyzed DAP5 expression levels in GC and adjacent tissues and its association with survival outcomes of GC patients using public databases. We collected paired samples of GC and adjacent tissues from 102 patients undergoing radical resection of GC in our hospital from June, 2012 to July, 2017, and analyzed the correlation of DAP5 expression level detected immunohistochemically with the clinicopathological parameters of the patients. Cox regression analysis, Kaplan-Meier analysis, and ROC curves were used to explore the independent risk factors and the predictive value of DAP5 expression for 5-year survival of the patients. In the cell experiments, we observed the changes in aerobic glycolysis in MGC-803 cells following lentivirus-mediated DAP5 knockdown or overexpression by measuring glucose uptake and cellular lactate level and using qRT-PCR and Western blotting. RESULTS: Analysis using the public databases showed that DAP5 was highly expressed in GC and correlated with tumor progression and poor survival outcomes of the patients (P < 0.05). In the clinical samples, DAP5 expression was significantly higher in GC than in the adjacent tissues (3.19±0.60 vs 1.00±0.12; t=36.863, P < 0.01), and a high expression of DAP5 was associated with a reduced 5-year survival rate of the patients (17.6% vs 72.5%; χ²=29.921, P < 0.05). A high DAP5 expression, T3-4, N2-3, and CEA≥5 ng/mL were identified as independent risk factors affecting 5-year survival outcomes of GC (P < 0.05), for which DAP5 expression showed a prediction sensitivity, specificity and accuracy of 73.2%, 80.4% and 79.0%, respectively. In MGC-803 cells, DAP5 knockdown significantly reduced glucose uptake, lactate level and the expressions of GLUT1, HK2 and LDHA, and DAP5 overexpression produced the opposite effects (P < 0.05). CONCLUSION A high expression of DAP5 in GC, which enhances cellular aerobic glycolysis to promote cancer progression, is correlated with a poor survival outcome and may serve as a biomarker for evaluating long-term prognosis of GC patients.
Humans ; Stomach Neoplasms ; Blotting, Western ; Databases, Factual ; Glucose ; Lactates

OBJECTIVE: To investigate the molecular mechanism underlying inherent fosfomycin resistance of Klebsiella pneumoniae (K. pneumoniae). METHODS: The draft genomic sequences of 14 clinical hypervirulent/hypermucoviscous K. pneumoniae (HvKP/ HmKP) isolates were obtained using the next-generation sequencing technology. The genomic sequences were analyzed using the Resistance Gene Identifier (RGI) software for predicting the resistome based on homology and SNP models in the Comprehensive Antibiotic Resistance Database (CARD) and for identification of the presence of phosphomycin resistancerelated genes uhpt and fosA and their mutations in the bacterial genomes. The results were verified by analyzing a total of 521 full-length genomic sequences of K. pneumonia strains obtained from GenBank. RESULTS: All the 14 clinical isolates of HvKP/ HmKP carried hexose phosphate transporter (UhpT) gene mutation, in which the glutamic acid was mutated to glutamine at 350aa (UhpT^E350Q mutation); the presence of fosA6 gene was detected in 12 (85.71%) of the isolates and fosA5 gene was detected in the other 2 (14.29%) isolates. Analysis of the genomic sequences of 521 K. pneumonia strains from GenBank showed that 508 (97.50%) strains carried UhpT^E350Q mutation, 439 (84.26%) strains harbored fosA6, and 80 (15.36%) strains harbored fosA5; 507 (97.31%) strains were found to have both UhpT^E350Q mutation and fosA6/5 genes in the genome. Only 12 (2.30%) strains carried fosA6/5 genes without UhpT^E350Q mutation; 1 (0.19%) strain had only UhpT^E350Q mutation without fosA6/5 genes, and another strain contained neither UhpT^E350Q mutation nor fosA6/5 genes. CONCLUSION UhpT^E350Q mutation with the presence of fosA6/5 genes are ubiquitous in K. pneumonia genomes, indicating a possible intrinsic mechanism of fosfomycin resistance in the bacterium to limit the use of fosfomycin against infections caused by K. pneumoniae, especially the multi-resistant HvKP/HmKP strains.
Fosfomycin ; Klebsiella pneumoniae ; Mutation ; Databases, Factual ; High-Throughput Nucleotide Sequencing

OBJECTIVE: To construct an inherent interpretability machine learning model as an explainable boosting machine model (EBM) for predicting one-year risk of death in patients with severe ischemic stroke. METHODS: We randomly divided the data of 2369 eligible patients with severe ischemic stroke in the MIMIC-Ⅳ(2.0) database, who were admitted in ICU in 2008 to 2019, into a training dataset (80%) and a test dataset (20%), and assessed the prognosis of the patients using the EBM model. The prediction performance of the model was evaluated by calculating the area under the receiver operating characteristic (AUC) curve. The calibration curve and Brier score were used to evaluate the degree of calibration of the model, and a decision curve was generated to assess the net clinical benefit. RESULTS: The EBM model constructed in this study had good discrimination power, calibration and net benefit, with an AUC of 0.857 (95% CI: 0.831-0.887) for predicting prognosis of severe ischemic stroke. Calibration curve analysis showed that the standard curve of the EBM model was the closest to the ideal curve. Decision curve analysis showed that the model had the greatest net benefit rate at the prediction probability threshold of 0.10 to 0.80. The top 5 independent predictive variables based on the EBM model were age, SOFA score, mean heart rate, mechanical ventilation, and mean respiratory rate, whose significance scores ranged from 0.179 to 0.370. CONCLUSION This EBM model has a good performance for predicting the risk of death within one year in patients with severe ischemic stroke and allows clinicians to better understand the contributing factors of the patients' outcomes through the model interpretability.
Humans ; Ischemic Stroke ; Calibration ; Databases, Factual ; Intensive Care Units ; Machine Learning

Addressing weight stigma is essential to obesity management as it causes inequalities in healthcare and impacts the outcomes of health. This narrative review summarises systematic review findings about the presence of weight bias in healthcare professionals, and interventions to reduce weight bias or stigma in these professionals. Two databases (PubMed and Cumulative Index to Nursing and Allied Health Literature [CINAHL]) were searched. Seven eligible reviews were identified from 872 search results. Four reviews identified the presence of weight bias, and three investigated trials to reduce weight bias or stigma in healthcare professionals. The findings may help further research and the treatment, health and well-being of individuals with overweight or obesity in Singapore. Weight bias was prevalent among qualified and student healthcare professionals globally, and there is a lack of clear guidance for effective interventions to reduce it, particularly in Asia. Future research is essential to identify the issues and inform initiatives to reduce weight bias and stigma among healthcare professionals in Singapore.
Humans ; Weight Prejudice ; Singapore ; Asia ; Databases, Factual ; Health Facilities