Leveraging Neural Networks to Profile Health Care Providers with Application to Medicare Claims (preprint)

Encompassing numerous nationwide, statewide, and institutional initiatives in the United States, provider profiling has evolved into a major health care undertaking with ubiquitous applications, profound implications, and high-stakes consequences. In line with such a significant profile, the literature has accumulated a number of developments dedicated to enhancing the statistical paradigm of provider profiling. Tackling wide-ranging profiling issues, these methods typically adjust for risk factors using linear predictors. While this approach is simple, it can be too restrictive to characterize complex and dynamic factor-outcome associations in certain contexts. One such example arises from evaluating dialysis facilities treating Medicare beneficiaries with end-stage renal disease. It is of primary interest to consider how the coronavirus disease (COVID-19) affected 30-day unplanned readmissions in 2020. The impact of COVID-19 on the risk of readmission varied dramatically across pandemic phases. To efficiently capture the variation while profiling facilities, we develop a generalized partially linear model (GPLM) that incorporates a neural network. Considering provider-level clustering, we implement the GPLM as a stratified sampling-based stochastic optimization algorithm that features accelerated convergence. Furthermore, an exact test is designed to identify under- and over-performing facilities, with an accompanying funnel plot to visualize profiles. The advantages of the proposed methods are demonstrated through simulation experiments and profiling dialysis facilities using 2020 Medicare claims from the United States Renal Data System.

CBPH assay for the highest sensitive detection of SARS-COV-2 in the semen.

BACKGROUND: Great concerns have been raised on SARS-CoV-2 impact on men's andrological well-being, and many studies have attempted to determine whether SARS-CoV-2 is present in the semen and till now the data are unclear and somehow ambiguous. However, these studies used quantitative real-time (qRT) PCR, which is not sufficiently sensitive to detect nucleic acids in clinical samples with a low viral load. METHODS: The clinical performance of various nucleic acid detection methods (qRT-PCR, OSN-qRT-PCR, cd-PCR, and CBPH) was assessed for SARS-CoV-2 using 236 clinical samples from laboratory-confirmed COVID-19 cases. Then, the presence of SARS-CoV-2 in the semen of 12 recovering patients was investigated using qRT-PCR, OSN-qRT-PCR, cd-PCR, and CBPH in parallel using 24 paired semen, blood, throat swab, and urine samples. RESULTS: The sensitivity and specificity along with AUC of CBPH was markedly higher than the other 3methods. Although qRT-PCR, OSN-qRT-PCR and cdPCR detected no SARS-CoV-2 RNA in throat swab, blood, urine, and semen samples of the 12 patients, CBPH detected the presence of SARS-CoV-2 genome fragments in semen samples, but not in paired urine samples, of 3 of 12 patients. The existing SARS-CoV-2 genome fragments were metabolized over time. CONCLUSIONS: Both OSN-qRT-PCR and cdPCR had better performance than qRT-PCR, and CBPH had the highest diagnostic performance in detecting SARS-CoV-2, which contributed the most improvement to the determination of the critical value in gray area samples with low vrial load, which then provides a rational screening strategy for studying the clearance of coronavirus in the semen over time in patients recovering from COVID-19. Although the presence of SARS-CoV-2 fragments in the semen was demonstrated by CBPH, COVID-19 is unlikely to be sexually transmitted from male partners for at least 3 months after hospital discharge.

Current status of management of medical waste in medical institutions of Chongqing

OBJECTIVE: To understand the status of generation and management of medical waste in medical institutions of Chongqing. METHODS: By means of onsite investigation and questionnaire survey, the generation categories and current status of management of medical waste in 50 medical institutions were investigated from Oct 2021 to Apr 2022 the existing limitations and prominent problems in the whole-process management of medical waste were identified so as to enable the safe disposal of medical waste based on laws and regulations. RESULTS: The average pollutants generation coefficient of medical waste was 0.22-0.72 kg/bed.day among all the grades of hospitals, the average pollutant generation coefficient of medical waste was 0.28-2.30 kg/10 people among grass-root medical institutions. The management of medical waste was more standardized in tertiary hospitals. There were a variety of problems in management of medical waste in clinics and village clinics, such as nonstandard classification of medical waste, unreasonable site selection for temporary storage of medical waste, unsatisfactory transportation means and untimely collection and transportation of medical waste. The problems of chemical, pharmaceutical and pathological medical waste were more prominent. The costs of disposal of medical waste were not strictly implemented in accordance with standards. The packaging, storage, loading, handover and disinfection of COVID-19 medical waste have been carried out in accordance with regulations. CONCLUSION: It is necessary to further standardize the management of medical waste, explore and formulate the collection and transportation modes of medical waste in primary medical institutions, intensify the supervision of classification, collection, storage, transportation and disposal of medical waste, optimize and upgrade the medical waste management information system, and encourage subsidies for the disposal of medical waste in Chongqing medical waste disposal enterprises during the COVID-19 period.

Mechanisms controlling the transport and evaporation of human exhaled respiratory droplets containing the severe acute respiratory syndrome coronavirus: a review.

Transmission of the coronavirus disease 2019 is still ongoing despite mass vaccination, lockdowns, and other drastic measures to control the pandemic. This is due partly to our lack of understanding on the multiphase flow mechanics that control droplet transport and viral transmission dynamics. Various models of droplet evaporation have been reported, yet there is still limited knowledge about the influence of physicochemical parameters on the transport of respiratory droplets carrying the severe acute respiratory syndrome coronavirus 2. Here we review the effects of initial droplet size, environmental conditions, virus mutation, and non-volatile components on droplet evaporation and dispersion, and on virus stability. We present experimental and computational methods to analyze droplet transport, and factors controlling transport and evaporation. Methods include thermal manikins, flow techniques, aerosol-generating techniques, nucleic acid-based assays, antibody-based assays, polymerase chain reaction, loop-mediated isothermal amplification, field-effect transistor-based assay, and discrete and gas-phase modeling. Controlling factors include environmental conditions, turbulence, ventilation, ambient temperature, relative humidity, droplet size distribution, non-volatile components, evaporation and mutation. Current results show that medium-sized droplets, e.g., 50 µm, are sensitive to relative humidity. Medium-sized droplets experience delayed evaporation at high relative humidity, and increase airborne lifetime and travel distance. By contrast, at low relative humidity, medium-sized droplets quickly shrink to droplet nuclei and follow the cough jet. Virus inactivation within a few hours generally occurs at temperatures above 40 °C, and the presence of viral particles in aerosols impedes droplet evaporation.

SARS-CoV-2-Related Acute Necrotizing Encephalopathy of Childhood With Good Response to Tocilizumab in an Adolescent.

BACKGROUND: Acute necrotizing encephalopathy of childhood (ANEC) is a rare parainfectious neurological disorder. ANEC is associated with a high mortality rate and poor neurological outcomes. ANEC is postulated to arise from immune-mediated or metabolic processes driven by viral infections. Although there have been some case reports of acute necrotizing encephalopathy with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) coinfection in adults, paediatric cases are rare. METHODS: A single case report of SARS-CoV-2-related ANEC in an 11-year-old boy is presented through retrospective chart review. Literature search was performed using PubMed, Embase, Cochrane database, and Google Scholar to compare and analyze similar cases of parainfectious immune-mediated encephalopathies related to SARS-CoV-2 in children. RESULTS: An 11-year-old boy with acute SARS-CoV-2 infection presented with ophthalmoplegia, ataxia, and aphasia. Neuroimaging findings demonstrated significant swelling and signal changes in bilateral thalami, brainstem, and cerebellar hemispheres, consistent with ANEC. His high ANEC Severity Score indicated poor neurological prognosis. Treatment with a combination of early steroid therapy, intravenous immunoglobulin therapy, and targeted interleukin 6 (IL-6) blockade yielded good neurological improvements. Literature search identified 19 parainfectious immune-mediated neurological disorders related to SARS-CoV-2 in children. The only other pediatric ANEC case identified was postinfectious and thus not included. CONCLUSIONS: This is the first report of a pediatric case of SARS-CoV-2-related ANEC, which responded well to early immunotherapy, including IL-6 blockade. Early immunotherapy with IL-6 blockade can be considered as an adjunct in managing severe ANEC.

Surgical workflow and infection prevention and control strategies for patients with COVID-19 during the epidemic in Children's Hospitals.

Background: This study sought to investigate the surgical workflow and practical effects of infection prevention and control in specialist pediatric hospitals during the Corona Virus Disease 2019 (COVID-19) epidemic to provide a foundation for ensuring the safety of children and medical staff. Methods: The Guidelines for the Management of Surgical Procedures and Infection Prevention and Control of COVID-19 Pneumonia in Children were formulated according to the industry specifications and standards, the prevention and control work system for hospitals in China, and the experiences of the Chinese Nursing Association in infection prevention and control in the operating room. These guidelines focus on the characteristics of children, and provide management priorities in relation to personnel management, infection prevention and control during surgery, intraoperative safety, and the cooperation of medical staff teams. These operation management and prevention and control strategies were applied to children who were suspected or confirmed to have COVID-19. Results: The operation process and prevention and control measures were effectively implemented. During the epidemic, a total of 219 surgeries which patients' COVID-19 nucleic test result are not out were completed. No medical staff or nosocomial infection occurred during the surgeries. Conclusions: As a special group, children are susceptible to COVID-19, and should receive special attention. As the only hospital designated to treat children with COVID-19 in Hubei Province, our hospital effectively implemented the infection prevention and control measures in surgery according to the characteristics of children. These measures ensured the safety of the surgeries and reduced the risk of infection in children and medical staff.

Rapid Point-of-Care Assay by SERS Detection of SARS-CoV-2 Virus and Its Variants.

Addressing the spread of coronavirus disease 2019 (COVID-19) has highlighted the need for rapid, accurate, and low-cost diagnostic methods that detect specific antigens for SARS-CoV-2 infection. Tests for COVID-19 are based on reverse transcription PCR (RT-PCR), which requires laboratory services and is time-consuming. Here, by targeting the SARS-CoV-2 spike protein, we present a point-of-care SERS detection platform that specifically detects SARS-CoV-2 antigen in one step by captureing substrates and detection probes based on aptamer-specific recognition. Using the pseudovirus, without any pretreatment, the SARS-CoV-2 virus and its variants were detected by a handheld Raman spectrometer within 5 min. The limit of detection (LoD) for the pseudovirus was 124 TU µL-1 (18 fM spike protein), with a linear range of 250-10,000 TU µL-1. Moreover, this assay can specifically recognize the SARS-CoV-2 antigen without cross reacting with specific antigens of other coronaviruses or influenza A. Therefore, the platform has great potential for application in rapid point-of-care diagnostic assays for SARS-CoV-2.

Trophic state assessment of optically diverse lakes using Sentinel-3-derived trophic level index

An accurate estimation of trophic state of lakes with satellite remote sensing is a challenge due to the optical complexity and variability associated with inland waters. Match-up data from 393 sampling stations that has concurrent Sentinel-3 OLCI images were acquired across Wuhan lakes. Trophic Level Index (TLI) algorithms were developed within a global Optical Water Type (OWT) classification system. The performance of algorithms with limited training data gathered by using spectral similarity of highest Sowt was not improved compared with that on basis of no classification. In contrast, using spectral similarity of Sowt > 0.9 rather than the highest Sowt to group more training data with similar traits for each OWT can help build more robust algorithms, which performance is better than that on basis of no classification. Algorithm performance statistics of the test dataset for the stepwise multiple linear regression (SMLR) method were the following: Mean Absolute Error (MAE) = 5.56;Mean Absolute Percentage Error (MAPE) = 11.02 %;Root Mean Square Error (RMSE) = 7.24 and for the back propagation neural network on the basis of the Levenberg-Marquardt-Bayesian regularization algorithm (LMBR-BPNN) method MAE = 4.56;MAPE = 8.33 %;RMSE = 5.98. We detected 8 different OWTs (2,3,4,5,9,10,11,12) in Wuhan lakes and clear spatio-temporal patterns of the trophic state between 2018 and 2020.Our results revealed that the trophic state of Wuhan lakes did not decrease as expected during the COVID-19 lockdown period.

Reduced B cell antigenicity of Omicron lowers host serologic response.

The SARS-CoV-2 Omicron variant evades most neutralizing vaccine-induced antibodies and is associated with lower antibody titers upon breakthrough infections than previous variants. However, the mechanism remains unclear. Here, we find using a geometric deep-learning model that Omicron's extensively mutated receptor binding site (RBS) features reduced antigenicity compared with previous variants. Mice immunization experiments with different recombinant receptor binding domain (RBD) variants confirm that the serological response to Omicron is drastically attenuated and less potent. Analyses of serum cross-reactivity and competitive ELISA reveal a reduction in antibody response across both variable and conserved RBD epitopes. Computational modeling confirms that the RBS has a potential for further antigenicity reduction while retaining efficient receptor binding. Finally, we find a similar trend of antigenicity reduction over decades for hCoV229E, a common cold coronavirus. Thus, our study explains the reduced antibody titers associated with Omicron infection and reveals a possible trajectory of future viral evolution.

Reduced B-cell antigenicity of Omicron lowers host serologic response

The SARS-CoV-2 Omicron variant evades most neutralizing vaccine-induced antibodies and is associated with lower antibody titers upon breakthrough infections than previous variants. However, the mechanism remains unclear. Here, we find using a geometric deep-learning model that Omicron's extensively mutated receptor binding site (RBS) features reduced antigenicity compared to previous variants. Mice immunization experiments with different recombinant Receptor Binding Domains (RBD) variants confirm that the serological response to Omicron is drastically attenuated and less potent. Analyses of serum cross-reactivity and competitive ELISA reveal a reduction in antibody response across both variable and conserved RBD epitopes. Computational modeling confirms that the RBS has a potential for further antigenicity reduction while retaining efficient receptor binding. Finally, we find a similar trend of antigenicity reduction over decades for hCoV229E, a common cold coronavirus. Thus our study explains the reduced antibody titers associated with Omicron infection and reveals a possible trajectory of future viral evolution. Graphical SARS-CoV-2 Omicron variant evades most neutralizing vaccine-induced antibodies and is associated with lower antibody titers upon breakthrough infections than previous variants. Tubiana et al. investigate the underlying mechanism using geometric deep learning, mice immunization experiments and biochemical assays. Mutations reduce antigenicity of the receptor binding site, leading to lower antibody response.

The Effect of Routine Management Combined with Case Management Model on Social Support, Self-Efficacy, Self-Management Ability, and Psychological Flexibility of AIDS Patients.

Objective: To explore the influence of conventional management combined with case management on social support and self-efficacy of AIDS patients. Methods: The clinical case data of 120 AIDS patients who were treated and nursed in our hospital from June 2019 to June 2021 were selected as the research objects and were divided into the control group and the observation group according to the digital table method, with 60 cases each. The control group implements routine management, and the observation group implements case-based nursing management on this basis and compares the effects of self-efficacy, self-management ability, nursing ability, social support, and psychological flexibility of the two groups of patients. Results: Before the intervention, the quality of life scores of the two groups was not statistically significant (P > 0.05). After the intervention, the physical function score, pain management score, and symptom response score of the observation group were significantly higher than those of the control group, and statistics showed that the difference was statistically significant (P < 0.05). Before the intervention, the self-management ability of the two groups of patients was not statistically significant (P > 0.05). After the intervention, the observation group's symptom management, emotional cognition management, social support and assistance, daily life management, disease knowledge management, and treatment compliance management were significantly higher than those of the control group. Statistics show that this difference is statistically significant (P < 0.05). Before the intervention, there was no significant difference in the nursing ability and psychological flexibility between the two groups of patients (P > 0.05). After the intervention, the observation group's health knowledge level, self-care skills, self-care responsibility, self-concept, and mental flexibility (resilience, strength, optimism) indicators were higher than the control group, while the depression mood disorder score was significantly lower than the control group; statistics showed that this difference was statistically significant (P < 0.05). Conclusion: Routine management combined with case-based nursing management can effectively improve the self-management ability and psychological flexibility of AIDS patients, improve patient care ability and self-efficacy, and provide certain reference value for effective management of AIDS patients.

Metabolite profile of COVID-19 revealed by UPLC-MS/MS-based widely targeted metabolomics

The metabolic characteristics of COVID-19 disease are still largely unknown. Here, 44 patients with COVID-19 (31 mild COVID-19 patients and 13 severe COVID-19 patients), 42 healthy controls (HC), and 42 patients with community-acquired pneumonia (CAP), were involved in the study to assess their serum metabolomic profiles. We used widely targeted metabolomics based on an ultra-performance liquid chromatography–tandem mass spectrometry (UPLC-MS/MS). The differentially expressed metabolites in the plasma of mild and severe COVID-19 patients, CAP patients, and HC subjects were screened, and the main metabolic pathways involved were analyzed. Multiple mature machine learning algorithms confirmed that the metabolites performed excellently in discriminating COVID-19 groups from CAP and HC subjects, with an area under the curve (AUC) of 1. The specific dysregulation of AMP, dGMP, sn-glycero-3-phosphocholine, and carnitine was observed in the severe COVID-19 group. Moreover, random forest analysis suggested that these metabolites could discriminate between severe COVID-19 patients and mild COVID-19 patients, with an AUC of 0.921. This study may broaden our understanding of pathophysiological mechanisms of COVID-19 and may offer an experimental basis for developing novel treatment strategies against it.

Reliability of Non-Contact Infrared Thermometers for Fever Screening Under COVID-19

Purpose Fever is one of the most typical clinical symptoms of coronavirus disease 2019 (COVID-19), and non-contact infrared thermometers (NCITs) are commonly used to screen for fever. However, there is a lack of authoritative data to define a “fever” when an NCIT is used and previous studies have shown that NCIT readings fluctuate widely depending on ambient temperatures and the body surface site screened. The aim of this study was to establish cut-off points for normal temperatures of different body sites (neck, forehead, temples, and wrist) and investigate the accuracy of NCITs at various ambient temperatures to improve the standardization and accuracy of fever screening. Patients and Methods A prospective investigation was conducted among 904 participants in the outpatient and emergency departments of Chengdu Women’s and Children’s Central Hospital. Body temperature was measured using NCITs and mercury axillary thermometers. A receiver operating characteristic curve was used to determine the accuracy of body temperature detection at the four body surface sites. Data on participant characteristics were also collected. Results Among the four surface sites, the neck temperature detection group had the highest accuracy. When the neck temperature was 37.35°C as the optimum fever diagnostic threshold, the sensitivity was 0.866. The optimum fever diagnostic thresholds for forehead, temporal, and wrist temperature were 36.65°C, 36.65°C, and 36.75°C, respectively. Moreover, triple neck temperature detection had the highest sensitivity, up to 0.998, whereas the sensitivity of triple wrist temperature detections was 0.949. Notably, the accuracy of NCITs significantly reduced when the temperature was lower than 18°C. Conclusion Neck temperature had the highest accuracy among the four NCIT temperature measurement sites, with an optimum fever diagnostic threshold of 37.35°C. Considering the findings reported in our study, we recommend triple neck temperature detection with NCITs as the fever screening standard for COVID-19.

[Characteristics, Ozone Formation Potential, and Source Apportionment of VOCs During the COVID-19 Epidemic in Xiong'an].

Many restrictive measures were implemented in China from January-February 2020 to control the rapid spread of COVID-19. Many studies reported that the COVID-19 lockdown impacted PM2.5, SO2, volatile organic compounds (VOCs), etc. VOCs play important roles in the production of ozone and PM2.5. Ambient VOCs in Xiong'an were measured from December 25, 2019 to January 24, 2020 (prior to epidemic prevention, P1) and from January 25, 2020 to February 24, 2020 (during epidemic prevention, P2) through a VOCs online instrument. In the study, VOCs characteristics and ozone generation potential (OFP) of ambient VOCs were analyzed, and source apportionment of VOCs were analyzed by using Positive Matrix Factorization (PMF). The results showed that φ(TVOCs) during epidemic prevention and control was 45.1×10-9, which was approximately half of that before epidemic prevention and control (90.5×10-9). The chemical composition of VOCs showed significant changes after epidemic prevention and control, the contribution rate of alkanes increased from 37.6% to 53.8%, and the contribution rate of aromatic hydrocarbons and halogenated hydrocarbons decreased from 13.3% and 12.0% to 7.5% and 7.8%, respectively. Aromatic hydrocarbons, halogenated hydrocarbons, and OVOCs decreased by more than 60%. Seven types of the top ten species were the same before and during the epidemic prevention and control, mainly low-carbon alkanes, olefins, aldehydes, and ketones. Dichloromethane, trichloromethane, and BTEXs decreased significantly. The OPP was 566 µg·m-3 and 231 µg·m-3 in P1 and P2, respectively. The OPP of VOCs decreased by more than 30%. The proportion of OFP contribution of aromatic hydrocarbons decreased significantly after the epidemic prevention and control, and the proportion of OFP contribution of alkanes and alkynes increased significantly. Positive matrix factorization (PMF) was then applied for VOCs sources apportionment. Six sources were identified, including background sources, oil-gas volatile sources, combustion sources, industrial sources, solvent use sources, and vehicle exhaust sources. The results showed that after the epidemic prevention and control, the contribution rate of solvent use sources to TVOCs decreased from 24% to 9%. The contribution rates of background sources, oil-gas volatile sources, and combustion sources increased from 13%, 34%, and 24% to 6%, 14%, and 13%, respectively. The relative contributions of vehicle exhaust sources before and after epidemic prevention and control were 21% and 18%, respectively. The observation points were affected by the emission of VOCs from paroxysmal industrial sources before the epidemic prevention and control. The emission was stopped after the epidemic prevention and control, and its contribution rate was reduced from 22% before the epidemic prevention and control to 1%. The concentrations of industrial sources, solvent sources, motor vehicle tail gas sources, and combustion sources decreased by 97%, 82%, 61%, and 15%, respectively, after the epidemic prevention and control. The concentration of background sources remained stable, and the concentration of oil and gas volatile sources increased by 7%. The control of production and traffic activities cannot reduce the emission of VOCs from oil and gas volatile sources, which is the focus of VOCs control in Xiong'an.

A Review of Deep Learning in Multiscale Agricultural Sensing

Population growth, climate change, and the worldwide COVID-19 pandemic are imposing increasing pressure on global agricultural production. The challenge of increasing crop yield while ensuring sustainable development of environmentally friendly agriculture is a common issue throughout the world. Autonomous systems, sensing technologies, and artificial intelligence offer great opportunities to tackle this issue. In precision agriculture (PA), non-destructive and non-invasive remote and proximal sensing methods have been widely used to observe crops in visible and invisible spectra. Nowadays, the integration of high-performance imagery sensors (e.g., RGB, multispectral, hyperspectral, thermal, and SAR) and unmanned mobile platforms (e.g., satellites, UAVs, and terrestrial agricultural robots) are yielding a huge number of high-resolution farmland images, in which rich crop information is compressed. However, this has been accompanied by challenges, i.e., ways to swiftly and efficiently making full use of these images, and then, to perform fine crop management based on information-supported decision making. In the past few years, deep learning (DL) has shown great potential to reshape many industries because of its powerful capabilities of feature learning from massive datasets, and the agriculture industry is no exception. More and more agricultural scientists are paying attention to applications of deep learning in image-based farmland observations, such as land mapping, crop classification, biotic/abiotic stress monitoring, and yield prediction. To provide an update on these studies, we conducted a comprehensive investigation with a special emphasis on deep learning in multiscale agricultural remote and proximal sensing. Specifically, the applications of convolutional neural network-based supervised learning (CNN-SL), transfer learning (TL), and few-shot learning (FSL) in crop sensing at land, field, canopy, and leaf scales are the focus of this review. We hope that this work can act as a reference for the global agricultural community regarding DL in PA and can inspire deeper and broader research to promote the evolution of modern agriculture.

Urban Heat Islands Significantly Reduced by COVID-19 Lockdown

The significant reduction in human activities during COVID-19 lockdown is anticipated to substantially influence urban climates, especially urban heat islands (UHIs). However, the UHIs variations during lockdown periods remain to be quantified. Based on the MODIS daily land surface temperature and the in-situ surface air temperature observations, we reveal a substantial decline in both surface and canopy UHIs over 300-plus megacities in China during lockdown periods compared with reference periods. The surface UHI intensity (UHII) is reduced by 0.25 (one S.D. = 0.22) K in the daytime and by 0.23 (0.20) K at night during lockdown periods. The reductions in canopy UHII reach 0.42 (one S.D. = 0.26) K in the daytime and 0.39 (0.29) K at night. These reductions are mainly due to the near-unprecedented drop in human activities induced by strict lockdown measures. Our results provide an improved understanding of the urban climate variations during the global pandemic.

Automatic CT Whole-Lung Segmentation in Radiomics Discrimination: Methodology and Application in Pneumonia Diagnosis and Distinguishment

Radiomics based on lesion segmentation has been widely accepted for disease diagnosis;however, it is difficult to precisely determine the boundary for pneumonia due to its diffuse characteristics. In this study, we aimed to propose an automatic radiomics method using whole-lung segmentation in pneumonia discrimination and assist clinical practitioners in fast and accurate diagnosis. In the discovery set, data from 151 participants diagnosed with type A or B influenza virus pneumonia, 63 diagnosed with coronavirus disease 2019 (COVID-19) and 50 healthy participants were collected. The three groups of data were compared in pairs. A total of 117 radiomics features were extracted from whole-lung images segmented by a four-layer U-net. We then utilized a logistic regression model to train the model and used the area under the receiver operating characteristic curve (AUC) to assess its performance. The L1 regularization term was used in feature selection, and 10-fold cross-validation was used to tune the hyperparameters. Fourteen radiomics features were selected to classify influenza pneumonia and health, and the AUC was 0.957 (95% confidential interval (CI): 0.939, 0.976) in the training set and 0.914 (95% CI: 0.866, 0.963) in the testing set. Eighteen features were selected for COVID-19 and health, and the AUC was 0.949 (95% CI: 0.926, 0.973)in the training set and 0.911 (95% CI: 0.859, 0.963) in the testing set. Twenty-eight features were selected for influenza virus pneumonia and COVID-19, and the AUC was 0.895 (95% CI: 0.870, 0.920) in the training set and 0.839 (95% CI: 0.791, 0.887) in the testing set. The results show that the automatic radiomics model based on whole lung segmentation is effective in distinguishing influenza virus pneumonia, COVID-19 and health, and may assist in the diagnosis of influenza virus pneumonia and COVID-19.

A Regional Industrial Economic Forecasting Model Based on a Deep Convolutional Neural Network and Big Data

To accurately predict the economic development of each industry in different types of regions, a deep convolutional neural network model was designed for predicting the annual GDP;GDP growth index;and primary, secondary and tertiary industry growth values of each. In the model, raw industrial data are preprocessed by a normalization operation and subsequently transformed by the BoxCox method to approach the normal distribution. Panel data of consecutive years are constructed and used as input to the deep convolutional neural network, and industrial data of year t + 1 are used as the output of the network. Simulation experiments were conducted to analyze 23 years of industrial economic data from 31 provinces, municipalities, and autonomous regions in China. The experimental results show that R-squared value is larger than 0.91 for all 31 provinces and root mean squared log errors (RMSLE) of all regions are less than 0.1, which demonstrate that the proposed method achieves high prediction accuracy with generalization capability and can accurately predict the economic growth trends of different types of regions.

Epidemiological Characteristics of Sporadic Nosocomial COVID-19 Infections From June 2020 to June 2021 in China: An Overview of Vaccine Breakthrough Infection Events.

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic has wreaked havoc on millions of people around the world. Although China quickly brought the Coronavirus disease (COVID-19) under control, there have been several sporadic outbreaks in different regions of China since June 2020. This article described the chronological nosocomial COVID-19 infection events related to several sporadic outbreaks of SARS-CoV-2 in different regions of China. We have reported epidemiological characteristics and management measures of sporadic nosocomial COVID-19 infections from June 2020 to June 2021 and specially focused on the domestic COVID-19 breakthrough infection in China, such as domestic COVID-19 breakthrough infection-a vaccinated healthcare professional working in the isolation ward of a designated COVID-19 hospital.