Deep learning based on biologically interpretable genome representation predicts two types of human adaptation of SARS-CoV-2 variants.

Explosively emerging SARS-CoV-2 variants challenge current nomenclature schemes based on genetic diversity and biological significance. Genomic composition-based machine learning methods have recently performed well in identifying phenotype-genotype relationships. We introduced a framework involving dinucleotide (DNT) composition representation (DCR) to parse the general human adaptation of RNA viruses and applied a three-dimensional convolutional neural network (3D CNN) analysis to learn the human adaptation of other existing coronaviruses (CoVs) and predict the adaptation of SARS-CoV-2 variants of concern (VOCs). A markedly separable, linear DCR distribution was observed in two major genes-receptor-binding glycoprotein and RNA-dependent RNA polymerase (RdRp)-of six families of single-stranded (ssRNA) viruses. Additionally, there was a general host-specific distribution of both the spike proteins and RdRps of CoVs. The 3D CNN based on spike DCR predicted a dominant type II adaptation of most Beta, Delta and Omicron VOCs, with high transmissibility and low pathogenicity. Type I adaptation with opposite transmissibility and pathogenicity was predicted for SARS-CoV-2 Alpha VOCs (77%) and Kappa variants of interest (58%). The identified adaptive determinants included D1118H and A570D mutations and local DNTs. Thus, the 3D CNN model based on DCR features predicts SARS-CoV-2, a major type II human adaptation and is qualified to predict variant adaptation in real time, facilitating the risk-assessment of emerging SARS-CoV-2 variants and COVID-19 control.

Association between adherence to a lifestyle behavior and the risk of asthma in overweight and obese adolescents.

With the increasing prevalence of overweight and obesity in children and adolescents, to actively prevent the occurrence of asthma in this population is important for reducing the burden of the disease. Lifestyle factors, including diet and exercise, are importance for overweight and obese adolescents, as well as an important modifiable factor affecting airway inflammation and asthma, whether healthy lifestyle was correlated with the risk of asthma in adolescents ≥ 12 years has not been reported. We suspected that there might be correlation between healthy lifestyle behaviors and the risk of asthma in overweight and obese adolescents. This cross-sectional study aimed to explore the association between the adherence to a healthy lifestyle behaviors and the risk of asthma in overweight and obese adolescents based on the data of 945 participants aged between 12-18 years from the National Health and Nutrition Examination Surveys (NHANES). Univariable and multivariable weighted Logistic regression models were applied to evaluate the association between healthy lifestyle behaviors with asthma risk in overweight and obese adolescents. Odds ratio (OR) and 95% confidence interval (CI) were applied as estimates. We found that the risk of asthma was reduced in overweight and obese adolescents with intermediate (OR = 0.40, 95%CI: 0.17-0.94) or good lifestyle behaviors (OR = 0.33, 95%CI: 0.13-0.86) in comparison to those with poor lifestyle behaviors. In summary, intermediate or good lifestyle behaviors was correlated with decreased risk of asthma in overweight and obese adolescents, which might provide a reference for making further prevention strategies for asthma in adolescents.

Fusion of Quality Evaluation Metrics and Convolutional Neural Network Representations for ROI Filtering in LC-MS.

Region of interest (ROI) extraction is a fundamental step in analyzing metabolomic datasets acquired by liquid chromatography-mass spectrometry (LC-MS). However, noises and backgrounds in LC-MS data often affect the quality of extracted ROIs. Therefore, developing effective ROI evaluation algorithms is necessary to eliminate false positives meanwhile keep the false-negative rate as low as possible. In this study, a deep fused filter of ROIs (dffROI) was proposed to improve the accuracy of ROI extraction by combining the handcrafted evaluation metrics with convolutional neural network (CNN)-learned representations. To evaluate the performance of dffROI, dffROI was compared with peakonly (CNN-learned representation) and five handcrafted metrics on three LC-MS datasets and a gas chromatography-mass spectrometry (GC-MS) dataset. Results show that dffROI can achieve higher accuracy, better true-positive rate, and lower false-positive rate. Its accuracy, true-positive rate, and false-positive rate are 0.9841, 0.9869, and 0.0186 on the test set, respectively. The classification error rate of dffROI (1.59%) is significantly reduced compared with peakonly (2.73%). The model-agnostic feature importance demonstrates the necessity of fusing handcrafted evaluation metrics with the convolutional neural network representations. dffROI is an automatic, robust, and universal method for ROI filtering by virtue of information fusion and end-to-end learning. It is implemented in Python programming language and open-sourced at https://github.com/zhanghailiangcsu/dffROI under BSD License. Furthermore, it has been integrated into the KPIC2 framework previously proposed by our group to facilitate real metabolomic LC-MS dataset analysis.

End Group Effect of Asymmetric Benzodithiophene-Based Donor with Liquid-Crystal State for Small-Molecule Binary Solar Cell.

Liquid-crystal small molecule donor (LC-SMD) is a new type organic semiconductor, which is attractive not only for the easy synthesis and purification, well-defined chemical structures, etc., but also for the LC state that makes the crystallinity and aggregation state of molecules adjustable. Here, one new LC-SMD (a-BTR-H4) is synthesized with 1D alkoxyl and 2D thiophene-alkylthiol side-chained benzo[1,2-b:4,5-b']dithiophene core, trithiophene π-bridge, and 3-(2-ethylhexyl) rhodanine end group. a-BTR-H4 shows low LC transition temperature, 117 °C, however, counterpart material (a-BTR-H5) with the same main structure but 3-ethyl rhodanine terminal group does not show LC properties. Although a-BTR-H4/H5 show similar Ultraviolet-visible absorption spectrum and energy levels, a-BTR-H4 affords relatively high photovoltaic performances due to favorable blend morphology produced by the consistent annealing temperature of Y6-based accepters and liquid crystal temperature of donors. Preliminary results indicate that a-BTR-H4 gains a power conversion efficiency (PCE) of 11.36% for Y6-based devices, which is ascribed to better light harvest as well as balanced carrier generation and transport, while a-BTR-H5 obtains 7.57% PCE. Therefore, some materials with unique nematic LC phase have great application potential in organic electronics, and further work to utilize a-BTR-H4 for high-performance device is underway.

Prediction of drug-likeness using graph convolutional attention network.

MOTIVATION: The drug-likeness has been widely used as a criterion to distinguish drug-like molecules from non-drugs. Developing reliable computational methods to predict the drug-likeness of compounds is crucial to triage unpromising molecules and accelerate the drug discovery process. RESULTS: In this study, a deep learning method was developed to predict the drug-likeness based on the graph convolutional attention network (D-GCAN) directly from molecular structures. Results showed that the D-GCAN model outperformed other state-of-the-art models for drug-likeness prediction. The combination of graph convolution and attention mechanism made an important contribution to the performance of the model. Specifically, the application of the attention mechanism improved accuracy by 4.0%. The utilization of graph convolution improved the accuracy by 6.1%. Results on the dataset beyond Lipinski's rule of five space and the non-US dataset showed that the model had good versatility. Then, the billion-scale GDB-13 database was used as a case study to screen SARS-CoV-2 3C-like protease inhibitors. Sixty-five drug candidates were screened out, most substructures of which are similar to these of existing oral drugs. Candidates screened from S-GDB13 have higher similarity to existing drugs and better molecular docking performance than those from the rest of GDB-13. The screening speed on S-GDB13 is significantly faster than screening directly on GDB-13. In general, D-GCAN is a promising tool to predict the drug-likeness for selecting potential candidates and accelerating drug discovery by excluding unpromising candidates and avoiding unnecessary biological and clinical testing. AVAILABILITY AND IMPLEMENTATION: The source code, model and tutorials are available at https://github.com/JinYSun/D-GCAN. The S-GDB13 database is available at https://doi.org/10.5281/zenodo.7054367. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Comparative effectiveness of psychotherapies in adults with posttraumatic stress disorder: a network meta-analysis of randomised controlled trials.

BACKGROUND: Evidence on the long-term comparative effectiveness of posttraumatic stress disorder (PTSD) psychotherapies in adults remains unknown. Therefore, we performed an extensive network meta-analysis of randomised controlled trials (RCTs) to determine the comparative effectiveness of psychotherapies for people diagnosed with PTSD. METHODS: A comprehensive search was conducted in Cochrane library, Embase, Medline-OVID, PubMed, Scopus, and Psych-Info until March 2021. Studies on the effectiveness of cognitive processing therapy (CPT), cognitive therapy (CT), eye movement desensitisation reprocessing (EMDR), narrative exposure therapy (NET), prolonged exposure (PE), cognitive behavioural therapy (CBT), present-centred therapy (PCT), brief eclectic psychotherapies (BEP), psychodynamic therapy (PDT) or combination therapies compared to no treatment (NT) or treatment as usual (TAU) in adults with PTSD were included. Frequentist and Bayesian approaches were used for analysis in R-software. RESULTS: We included 98 RCTs with 5567 participants from 18 897 studies. CPT, EMDR, CT, NET, PE, CBT, and PCT were significant to reduce PTSD symptoms (SMD range: -1.53 to -0.75; Certainty: very low to high) at immediate post-treatment and ranked accordingly. Longitudinal analysis found EMDR (1.02) and CPT (0.85) as the significant therapies with large effect size in short-term and long-term follow-up, respectively. NET and CPT showed higher proportion of loss of PTSD diagnosis (RR range: 5.51-3.45) while there were no significant psychotherapies for retention rate compared to NT. CONCLUSIONS: Our findings provide evidence for improving current guidelines and informing clinical decision-making for PTSD management. However, the best PTSD treatment plan should be tailored to patients' needs, characteristics, and clinician expertise. REGISTRATION: PROSPERO CRD42020162143.

Association of dietary mineral mixture with depressive symptoms: A combination of Bayesian approaches.

The relationships between mixtures of multiple minerals and depression have not been explored. Therefore, we analyzed the relationship between the mixture of nine dietary minerals [calcium (Ca), phosphorus, magnesium (Mg), iron (Fe), zinc, copper (Cu), sodium, potassium (K), and selenium (Se)] and depressive symptoms in the general population. We screened 20,342 participants from the National Health and Nutrition Examination Survey (NHANES) 2007-2018. We fitted the general linear regression, Bayesian kernel machine regression (BKMR), and Bayesian semiparametric regression models to explore associations and interactions. We obtained the relative importance of dietary minerals by calculating posterior inclusion probabilities (PIPs). The dietary intakes of minerals were obtained using the 24-h dietary recall interview, and depressive symptoms were assessed using the Patient Health Questionnaire-9 (PHQ-9). The linear analysis showed that nine minerals were negatively associated with PHQ-9 scores. The BKMR analysis showed a negative association between the dietary mineral mixture and PHQ-9 scores, with Se having the largest PIP at 1.0000, followed by K (0.7784). We also observed potential interactions between Ca and Fe, Se and Fe, and K and Mg. Among them, the interaction of Ca and Fe had the largest PIP of 0.986. In addition, the overall effect was more pronounced in females than males, and Cu's PIP (0.8376) was higher in females. Two sensitivity analyses showed that our results were robust. Our study provides a basis for formulating nutritional intervention programs for depression in the future.

Efficacy of Blue LED Phototherapy on Sleep Quality and Behavioral and Psychological Symptoms of Dementia: A Double-Blind Randomized Controlled Trial.

INTRODUCTION: People with dementia often experience behavioral and psychological symptoms of dementia (BPSD), which are a major cause of caregiver burden and institutionalization. Therefore, we conducted a double-blind, parallel-group randomized controlled trial to examine the efficacy of blue-enriched light therapy for BPSD in institutionalized older adults with dementia. METHODS: Participants were enrolled and randomly allocated into blue-enriched light therapy (N = 30) or the conventional light group (N = 30) for 60 min in 10 weeks with five sessions per week. The primary outcome was sleep quality measured by actigraphy and Pittsburgh Sleep Quality Index (PSQI). The secondary outcome was overall BPSD severity (Cohen-Mansfield Agitation Inventory [CMAI] and Neuropsychiatric Inventory [NPI-NH]). The outcome indicators were assessed at baseline, mid-test, immediate posttest, 1-month, 3-month, and 6-month follow-up. The effects of the blue-enriched light therapy were examined by the generalized estimating equation model. RESULTS: Blue-enriched light therapy revealed significant differences in the objective sleep parameters (sleep efficiency: ß = 5.81, Waldχ2 = 32.60, CI: 3.82; 7.80; sleep latency: ß = -19.82, Waldχ2 = 38.38, CI:-26.09; -13.55), subjective sleep quality (PSQI: ß = -2.07, Waldχ2 = 45.94, CI: -2.66; -1.47), and overall BPSD severity (CMAI: ß = -0.90, Waldχ2 = 14.38, CI: -1.37; -0.44) (NPI-NH: ß = -1.67, Waldχ2 = 30.61, CI: -2.26; -1.08) compared to conventional phototherapy immediate posttest, 1-month, 3-month, and 6-month follow-up. Furthermore, the effects for sleep efficiency and sleep latency lasted for up to 6 months. In the subscale analysis, the differences of the behavioral symptoms changed significantly between the groups in physical/nonaggressive (CI: -1.01; -0.26) and verbal/nonaggressive (CI: -0.97; -0.29). CONCLUSIONS: Blue-enriched light therapy is a feasible low-cost intervention that could be integrated as a comprehensive therapy program for BPSD among older adults with dementia.

Genetic and Environmental Influences on Serum Alanine Aminotransferase Level: A Chinese Twin Study.

An abnormal alanine aminotransferase (ALT) level is predictive of disease and all-cause mortality and may indicate liver injury. Using twin modeling, the genetic and environmental factors that affect human serum ALT levels have been well studied for the populations in the different countries, and the results showed moderate-to-high heritability. However, the heritability of ALT level has not been explored in Chinese population. Thus, we recruited 369 pairs of twins (233 monozygotic and 136 dizygotic) from the Qingdao Twin Registry in China with a median age of 50 years (40-80 years). Correlation analysis and a structural equation model (SEM) were conducted to evaluate the heritability of ALT level. The data for age, gender, body mass index and alcohol consumption were set as covariates. Intrapair correlation in monozygotic twins was 0.64 (95%CI [.56, .71]) and 0.42 (95% CI [.28, .55]) in dizygotic twins. The SEM analysis indicated that 65% (95% CI [57%, 71%]) of the variation in ALT levels can be explained by additive genetics and 35% (95% CI [29%, 44%]) of the variation is attributed to unique environmental factors or residuals. Shared environmental influences were not significant. In conclusion, serum ALT variations exhibited strong genetic effects. The variation could also be explained by unique environmental factors. However, shared environmental factors have a minor impact on the serum ALT level.

A Robust Planar Marker-Based Visual SLAM.

Many visual SLAM systems are generally solved using natural landmarks or optical flow. However, due to textureless areas, illumination change or motion blur, they often acquire poor camera poses or even fail to track. Additionally, they cannot obtain camera poses with a metric scale in the monocular case. In some cases (such as when calibrating the extrinsic parameters of camera-IMU), we prefer to sacrifice the flexibility of such methods to improve accuracy and robustness by using artificial landmarks. This paper proposes enhancements to the traditional SPM-SLAM, which is a system that aims to build a map of markers and simultaneously localize the camera pose. By placing the markers in the surrounding environment, the system can run stably and obtain accurate camera poses. To improve robustness and accuracy in the case of rotational movements, we improve the initialization, keyframes insertion and relocalization. Additionally, we propose a novel method to estimate marker poses from a set of images to solve the problem of planar-marker pose ambiguity. Compared with the state-of-art, the experiments show that our system achieves better accuracy in most public sequences and is more robust than SPM-SLAM under rotational movements. Finally, the open-source code is publicly available and can be found at GitHub.

[Nursing cooperation in surgical collection of testis tissues from prepubertal male patients for cryopreservation: A study of 4 cases].

OBJECTIVE: To explore nursing cooperation in surgical collection of the testis tissue from prepubertal male patients for cryopreservation. METHODS: We retrospectively analyzed the methods and effects of perioperative nursing in surgical collection of the testis tissue from 4 prepubertal male patients for cryopreservation in our Center of Reproductive Medicine. Before, during and after operation, we took strict measures in making sterilized ice containers, intraoperative nursing cooperation, protection of the isolated testis tissues and transferring of the samples. RESULTS: Testis tissues were successfully collected from all the 4 prepubertal males, 31, 31, 20 and 34 samples from each case respectively, well protected and subjected to slow cryopreservation after standard processing in the embryo laboratory. CONCLUSION: In surgical collection of the testis tissue for cryopreservation, preparation of sterilized ice containers, intraoperative nursing cooperation and protection and transferring of the samples are essential for standard processing and cryopreservation of the testis tissue in the embryo laboratory.

Research Progress on Microbial Community Succession in the Postmortem Interval Estimation.

The postmortem interval (PMI) estimation is a key and difficult point in the practice of forensic medicine, and forensic scientists at home and abroad have been searching for objective, quantifiable and accurate methods of PMI estimation. With the development and combination of high-throughput sequencing technology and artificial intelligence technology, the establishment of PMI model based on the succession of the microbial community on corpses has become a research focus in the field of forensic medicine. This paper reviews the technical methods, research applications and influencing factors of microbial community in PMI estimation explored by using high-throughput sequencing technology, to provide a reference for the related research on the use of microbial community to estimate PMI.

ACSL4 is overexpressed in psoriasis and enhances inflammatory responses by activating ferroptosis.

More and more studies have shown that ferroptosis is closely related to the progression of various diseases, but the significance of ferroptosis in psoriasis is still rarely explored. The detection of plasma and psoriatic lesions found that the contents of MDA and ROS were significantly increased, while the contents of SOD and GSH were significantly decreased, and the trend of increase or decrease in patients with progressive psoriasis was more obvious. The expression of ACSL4, a key regulator of ferroptosis, was significantly increased in psoriatic lesions and further up-regulated in patients with progressive psoriasis. ACSL4 expression was positively correlated with PASI score and the expression levels of inflammatory cytokines (TNF-α, IL-6, IL-8 and IL-17a), and linear regression analysis showed that high expression of ACSL4 in psoriatic lesions was associated with higher PASI score. Both ferroptosis inducer Erastin and IFN-γ/TNF-α significantly induced ferroptosis, inhibited keratinocyte viability, promoted the accumulation of MDA, ROS and Fe2+, and enhanced ACSL4, TNF-α, IL-6 and IL-8 expression. When ferroptosis inhibitor Ferrostatin-1 was added to inhibit ferroptosis, the up-regulation trends of MDA, ROS, Fe2+, ACSL4, TNF-α, IL-6 and IL-8 were significantly inhibited, and inhibition of ACSL4 expression also had a similar effect. Apoptosis inhibitor Z-VAD-FMK could also attenuate the pro-inflammatory effect of IFN-γ/TNF-α, and Fer-1 plus Z-VAD-FMK further inhibited the expression of inflammatory cytokines. Thus, ferroptosis is significantly activated during the progression of psoriasis and promotes inflammatory responses by upregulating ACSL4 expression. This discovery will provide new targets for clinical detection, prevention and treatment of psoriasis.

Association of annual hospital septic shock case volume and hospital mortality.

BACKGROUND: The burden of sepsis remains high in China. The relationship between case volume and hospital mortality among patients with septic shock, the most severe complication of sepsis, is unknown in China. METHODS: In this retrospective cohort study, we analyzed surveillance data from a national quality improvement program in intensive care units (ICUs) in China in 2020. Association between septic shock case volume and hospital mortality was analyzed using multivariate linear regression and restricted cubic splines. RESULTS: We enrolled a total of 134,046 septic shock cases in ICUs from 1902 hospitals in China during 2020. In this septic shock cohort, the median septic shock volume per hospital was 33 cases (interquartile range 14-76 cases), 41.4% were female, and more than half of the patients were over 61 years old, with average hospital mortality of 21.2%. An increase in case volume was associated with improved survival among septic shock cases. In the linear regression model, the highest quartile of septic shock volume was associated with lower hospital mortality compared with the lowest quartile (ß - 0.86; 95% CI - 0.98, - 0.74; p < 0.001). Similar differences were found in hospitals of respective geographic locations and hospital levels. With case volume modeled as a continuous variable in a restricted cubic spline, a lower volume threshold of 40 cases before which a substantial reduction of the hospital mortality rate was observed. CONCLUSIONS: The findings suggest that hospitals with higher septic shock case volume have lower hospital mortality in China. Further research is needed to explain the mechanism of this volume-outcome relationship.

Blood retinol and retinol-binding protein concentrations are associated with diabetes: a systematic review and meta-analysis of observational studies.

PURPOSE: The associations between blood retinol, retinol-binding protein (RBP) concentrations and diabetes mellitus were inconsistent in literature. The objective is to investigate these associations by a systematic review and meta-analysis and provide basis for clinical intervention. METHODS: PubMed, Web of science, and Cochrane databases were searched from the beginning to July 1, 2021. A total of 13 studies on retinol and 31 studies on RBP are included in the current meta-analysis. RESULTS: The blood retinol concentration was significantly lower in the type I diabetes mellitus (T1DM) [standardized mean difference (SMD) (95% CI): - 0.59 (- 0.81, - 0.37), P < 0.01] and gestational diabetes mellitus (GDM) patients [SMD (95% CI): - 0.54 (- 0.87, - 0.20), P < 0.01] than in the controls. However, the difference was not significant between the type II diabetes mellitus (T2DM) patients and the controls. The RBP concentration was significantly higher in the diabetic patients than in the controls [SMD (95% CI): 0.24 (0.12, 0.35), P < 0.01]. Particularly, the RBP concentration was significantly higher in the T2DM and GDM patients. CONCLUSION: The blood retinol concentration was negatively associated with T1DM and GDM, while the blood RBP concentration was positively associated with T2DM and GDM. Future work should use a more sensitive retinol measurement method like retinol isotope dilution method to confirm whether blood retinol concentration differs between the diabetes patients and the controls.

Prevalence of grief disorders in bereaved families of cancer patients: A meta-analysis.

BACKGROUND: Cancer caregiving is a distressing experience and loss of a loved one can lead to intense grief and other adverse effects. However, the prevalence of grief disorders among families associated with cancer-related deaths remains unknown. AIM: This study aimed to determine the prevalence of grief disorders among families of patients with cancer to better inform clinicians, researchers, and policymakers. DESIGN: Meta-analysis, PROSPERO number CRD42020209392. DATA SOURCES: The databases of CINAHL, Embase, MEDLINE, PubMed, Scopus, PsycINFO, and Web of Science were comprehensively searched with no language restrictions. The quality of included studies was assessed with Hoy's criteria. RESULTS: Among the 3046 records screened, 19 studies were eligible for meta-analysis, with a total of 14,971 participants. The pooled prevalence rate of grief disorders was estimated at 14.2% (95% CI, 11.7%-16.7%), ranging from 7% to 39%. The prevalence was higher in females (10%; 95% CI, 8.2%-12.1%), those who are religious (9.55%; 95% CI, 8.97%-10.16%), spouses of the deceased (7.78%; 95% CI, 6.08%-9.69%), and families of patients with neurological cancers (6.4%; 95% CI, 0.10%-19.9%). Educational levels, study locations, diagnosis tools, time post-after loss, and study methods seemed not to affect the prevalence of grief disorders in families of patients with cancer. CONCLUSIONS: As the prevalence of grief disorders in cancer-related bereavement is substantial, therefore, support including palliative care is important to reduce the burden of caregiving. In addition, future studies are needed to identify and explore effective strategies that can help reduce the burden caused by grief disorders after the death of the patient.

Prevalence of Oropharyngeal Dysphagia and Risk of Pneumonia, Malnutrition, and Mortality in Adults Aged 60 Years and Older: A Meta-Analysis.

INTRODUCTION: Oropharyngeal dysphagia (OD) is a serious health condition associated with poor survival and quality of life in adults aged 60 years and older. Comprehensive assessment and management of OD could lead to better and improved health outcomes for older adults. Therefore, we performed the first meta-analysis to determine the pooled prevalence of OD and risk of pneumonia, malnutrition, and mortality. METHODS: Databases including Ovid-MEDLINE, Web of Science, Embase, PubMed, Cochrane, and CINAHL were searched up to January 2021. Data analysis was conducted using logistic-normal for prevalence rate and DerSimonian-Lard random-effects models for outcomes and associated factors of OD, providing odds ratio (OR) and corresponding 95% confidence interval (CI). RESULTS: The pooled prevalence of OD in 39 studies with 31,488 participants was 46% associated with higher pooled OR for risk of pneumonia 2.07 (95% CI, 1.58-2.72), malnutrition 2.21 (95% CI, 1.43-3.41), and mortality 2.73 (95% CI, 1.62-4.60). Geriatric syndromes including fecal incontinence 6.84 (4.955-9.44), immobility syndrome 6.06 (5.28-6.96), pressure ulcers 4.02 (2.46-6.56), sarcopenia 3.10 (1.89-5.09), urinary incontinence 2.75 (1.81-4.19), frailty 2.66 (1.16-6.13), delirium 2.23 (1.73-2.87), and falls 1.47 (1.19-1.81) and comorbidities including dementia 3.69 (2.36-5.78) and stroke 1.92 (1.47-2.52) were associated with OD. CONCLUSION: Early identification and management of OD should consider geriatric syndromes and neurogenic comorbidities to prevent malnutrition and pneumonia and reduce mortality in adults aged 60 years and older.

Prevalence of dysphagia and risk of pneumonia and mortality in acute stroke patients: a meta-analysis.

BACKGROUND: Post-stroke dysphagia (PSD) has been associated with high risk of aspiration pneumonia and mortality. However, limited evidence on pooled prevalence of post-stroke dysphagia and influence of individual, disease and methodological factors reveals knowledge gap. Therefore, to extend previous evidence from systematic reviews, we performed the first meta-analysis to examine the pooled prevalence, risk of pneumonia and mortality and influence of prognostic factors for PSD in acute stroke. METHODS: Our search was conducted in CINAHL, Cochrane Library, EMBASE, Ovid-Medline, PubMed, and Web of Science an initial search in October 2020 and a follow-up search in May 2021. Data synthesis was conducted using the Freeman-Tukey double-arcsine transformation model for the pooled prevalence rate and the DerSimonian-Lard random-effects model for prognostic factors and outcomes of PSD. RESULTS: The pooled prevalence of PSD was 42% in 42 studies with 26,366 participants. PSD was associated with higher pooled odds ratio (OR) for risk of pneumonia 4.08 (95% CI, 2.13-7.79) and mortality 4.07 (95% CI, 2.17-7.63). Haemorrhagic stroke 1.52 (95% CI, 1.13-2.07), previous stroke 1.40 (95% CI, 1.18-1.67), severe stroke 1.38 (95% CI, 1.17-1.61), females 1.25 (95% CI, 1.09-1.43), and diabetes mellitus 1.24 (95% CI, 1.02-1.51) were associated with higher risk of PSD. Males 0.82 (95% CI, 0.70-0.95) and ischaemic stroke 0.54 (95% CI, 0.46-0.65) were associated with lower risk of PSD. Haemorrhagic stroke, use of instrumental assessment method, and high quality studies demonstrated to have higher prevalence of PSD in the moderator analysis. CONCLUSIONS: Assessment of PSD in acute stroke with standardized valid and reliable instruments should take into account stroke type, previous stroke, severe stroke, diabetes mellitus and gender to aid in prevention and management of pneumonia and thereby, reduce the mortality rate. TRIAL REGISTRATION: https://osf.io/58bjk/?view_only=26c7c8df8b55418d9a414f6d6df68bdb .

Dynamic prediction of life-threatening events for patients in intensive care unit.

BACKGROUND: Early prediction of patients' deterioration is helpful in early intervention for patients at greater risk of deterioration in Intensive Care Unit (ICU). This study aims to apply machine learning approaches to heterogeneous clinical data for predicting life-threatening events of patients in ICU. METHODS: We collected clinical data from a total of 3151 patients admitted to the Medical Intensive Care Unit of Peking Union Medical College Hospital in China from January 1st, 2014, to October 1st, 2019. After excluding the patients who were under 18 years old or stayed less than 24 h at the ICU, a total of 2170 patients were enrolled in this study. Multiple machine learning approaches were utilized to predict life-threatening events (i.e., death) in seven 24-h windows (day 1 to day 7) and their performance was compared. RESULTS: Light Gradient Boosting Machine showed the best performance. We found that life-threatening events during the short-term windows can be better predicted than those in the medium-term windows. For example, death in 24 h can be predicted with an Area Under Curve of 0.905. Features like infusion pump related fluid input were highly related to life-threatening events. Furthermore, the prediction power of static features such as age and cardio-pulmonary function increased with the extended prediction window. CONCLUSION: This study demonstrates that the integration of machine learning approaches and large-scale high-quality clinical data in ICU could accurately predict life-threatening events for ICU patients for early intervention.

Deep Learning for Joint Pilot Design and Channel Estimation in MIMO-OFDM Systems.

In MIMO-OFDM systems, pilot design and estimation algorithm jointly determine the reliability and effectiveness of pilot-based channel estimation methods. In order to improve the channel estimation accuracy with less pilot overhead, a deep learning scheme for joint pilot design and channel estimation is proposed. This new hybrid network structure is named CAGAN, which is composed of a concrete autoencoder (concrete AE) and a conditional generative adversarial network (cGAN). We first use concrete AE to find and select the most informative position in the time-frequency grid to achieve pilot optimization design and then input the optimized pilots to cGAN to complete channel estimation. Simulation experiments show that the CAGAN scheme outperforms the traditional LS and MMSE estimation methods with fewer pilots, and has good robustness to environmental noise.