Predicting Mortality in Intensive Care Unit Patients With Heart Failure Using an Interpretable Machine Learning Model: Retrospective Cohort Study.

BACKGROUND: Heart failure (HF) is a common disease and a major public health problem. HF mortality prediction is critical for developing individualized prevention and treatment plans. However, due to their lack of interpretability, most HF mortality prediction models have not yet reached clinical practice. OBJECTIVE: We aimed to develop an interpretable model to predict the mortality risk for patients with HF in intensive care units (ICUs) and used the SHapley Additive exPlanation (SHAP) method to explain the extreme gradient boosting (XGBoost) model and explore prognostic factors for HF. METHODS: In this retrospective cohort study, we achieved model development and performance comparison on the eICU Collaborative Research Database (eICU-CRD). We extracted data during the first 24 hours of each ICU admission, and the data set was randomly divided, with 70% used for model training and 30% used for model validation. The prediction performance of the XGBoost model was compared with three other machine learning models by the area under the curve. We used the SHAP method to explain the XGBoost model. RESULTS: A total of 2798 eligible patients with HF were included in the final cohort for this study. The observed in-hospital mortality of patients with HF was 9.97%. Comparatively, the XGBoost model had the highest predictive performance among four models with an area under the curve (AUC) of 0.824 (95% CI 0.7766-0.8708), whereas support vector machine had the poorest generalization ability (AUC=0.701, 95% CI 0.6433-0.7582). The decision curve showed that the net benefit of the XGBoost model surpassed those of other machine learning models at 10%~28% threshold probabilities. The SHAP method reveals the top 20 predictors of HF according to the importance ranking, and the average of the blood urea nitrogen was recognized as the most important predictor variable. CONCLUSIONS: The interpretable predictive model helps physicians more accurately predict the mortality risk in ICU patients with HF, and therefore, provides better treatment plans and optimal resource allocation for their patients. In addition, the interpretable framework can increase the transparency of the model and facilitate understanding the reliability of the predictive model for the physicians.

Clinical characteristics and outcomes of lung cancer patients with COVID-19: A systematic review and meta-analysis protocol.

BACKGROUND: COVID-19 is spreading rapidly worldwide, and the population is generally susceptible to SARS-CoV-2, especially those with cancer. Hence, our study aims to design a protocol for a systematic review and meta-analysis of the clinical characteristics and prognoses of lung cancer patients with COVID-19. METHODS: The protocol is prepared following the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines. The literature will be searched in Embase, Pubmed, the Cochrane Library, LitCovid, and CNKI for potentially eligible articles. The quality of the articles will be used in the Newcastle-Ottawa Quality Assessment Scale (NOS) and Cochrane Handbook for Systematic Reviews of Interventions. Statistical analysis will be performed through RevMan 5 software. This review protocol has been registered in PROSPERO (CRD42022306866). DISCUSSION: To clarify whether COVID-19 affects the clinical symptoms and prognoses of lung cancer patients. Further study is needed to establish the best evidence-based for the management of lung cancer patients with COVID-19. CONCLUSION: The definitive conclusion will be important to physicians effectively manage lung cancer patients with COVID-19.

Public attitudes toward COVID-19 vaccines on English-language Twitter: A sentiment analysis.

OBJECTIVE: To identify themes and temporal trends in the sentiment of COVID-19 vaccine-related tweets and to explore variations in sentiment at world national and United States state levels. METHODS: We collected English-language tweets related to COVID-19 vaccines posted between November 1, 2020, and January 31, 2021. We applied the Valence Aware Dictionary and sEntiment Reasoner tool to calculate the compound score to determine whether the sentiment mentioned in each tweet was positive (compound ≥ 0.05), neutral (-0.05 < compound < 0.05), or negative (compound ≤ -0.05). We applied the latent Dirichlet allocation analysis to extract main topics for tweets with positive and negative sentiment. Then we performed a temporal analysis to identify time trends and a geographic analysis to explore sentiment differences in tweets posted in different locations. RESULTS: Out of a total of 2,678,372 COVID-19 vaccine-related tweets, tweets with positive, neutral, and negative sentiments were 42.8%, 26.9%, and 30.3%, respectively. We identified five themes for positive sentiment tweets (trial results, administration, life, information, and efficacy) and five themes for negative sentiment tweets (trial results, conspiracy, trust, effectiveness, and administration). On November 9, 2020, the sentiment score increased significantly (score = 0.234, p = 0.001), then slowly decreased to a neutral sentiment in late December and was maintained until the end of January. At the country level, tweets posted in Brazil had the lowest sentiment score of -0.002, while tweets posted in the United Arab Emirates had the highest sentiment score of 0.162. The overall average sentiment score for the United States was 0.089, with Washington, DC having the highest sentiment score of 0.144 and Wyoming having the lowest sentiment score of 0.036. CONCLUSIONS: Public sentiment on COVID-19 vaccines varied significantly over time and geography. Sentiment analysis can provide timely insights into public sentiment toward the COVID-19 vaccine and guide public health policymakers in designing locally tailored vaccine education programs.

Leveraging Transfer Learning to Analyze Opinions, Attitudes, and Behavioral Intentions Toward COVID-19 Vaccines: Social Media Content and Temporal Analysis.

BACKGROUND: The COVID-19 vaccine is considered to be the most promising approach to alleviate the pandemic. However, in recent surveys, acceptance of the COVID-19 vaccine has been low. To design more effective outreach interventions, there is an urgent need to understand public perceptions of COVID-19 vaccines. OBJECTIVE: Our objective was to analyze the potential of leveraging transfer learning to detect tweets containing opinions, attitudes, and behavioral intentions toward COVID-19 vaccines, and to explore temporal trends as well as automatically extract topics across a large number of tweets. METHODS: We developed machine learning and transfer learning models to classify tweets, followed by temporal analysis and topic modeling on a dataset of COVID-19 vaccine-related tweets posted from November 1, 2020 to January 31, 2021. We used the F1 values as the primary outcome to compare the performance of machine learning and transfer learning models. The statistical values and P values from the Augmented Dickey-Fuller test were used to assess whether users' perceptions changed over time. The main topics in tweets were extracted by latent Dirichlet allocation analysis. RESULTS: We collected 2,678,372 tweets related to COVID-19 vaccines from 841,978 unique users and annotated 5000 tweets. The F1 values of transfer learning models were 0.792 (95% CI 0.789-0.795), 0.578 (95% CI 0.572-0.584), and 0.614 (95% CI 0.606-0.622) for these three tasks, which significantly outperformed the machine learning models (logistic regression, random forest, and support vector machine). The prevalence of tweets containing attitudes and behavioral intentions varied significantly over time. Specifically, tweets containing positive behavioral intentions increased significantly in December 2020. In addition, we selected tweets in the following categories: positive attitudes, negative attitudes, positive behavioral intentions, and negative behavioral intentions. We then identified 10 main topics and relevant terms for each category. CONCLUSIONS: Overall, we provided a method to automatically analyze the public understanding of COVID-19 vaccines from real-time data in social media, which can be used to tailor educational programs and other interventions to effectively promote the public acceptance of COVID-19 vaccines.

Physicians' Perspectives of Telemedicine During the COVID-19 Pandemic in China: Qualitative Survey Study.

BACKGROUND: Generalized restriction of movement due to the COVID-19 pandemic, together with unprecedented pressure on the health system, has disrupted routine care for non-COVID-19 patients. Telemedicine should be vigorously promoted to reduce the risk of infections and to offer medical assistance to restricted patients. OBJECTIVE: The purpose of this study was to understand physicians' attitudes toward and perspectives of telemedicine during and after the COVID-19 pandemic, in order to provide support for better implementation of telemedicine. METHODS: We surveyed all physicians (N=148), from October 17 to 25, 2020, who attended the clinical informatics PhD program at West China Medical School, Sichuan University, China. The physicians came from 57 hospitals in 16 provinces (ie, municipalities) across China, 54 of which are 3A-level hospitals, two are 3B-level hospitals, and one is a 2A-level hospital. RESULTS: Among 148 physicians, a survey response rate of 87.2% (129/148) was attained. The average age of the respondents was 35.6 (SD 3.9) years (range 23-48 years) and 67 out of 129 respondents (51.9%) were female. The respondents come from 37 clinical specialties in 55 hospitals located in 14 provinces (ie, municipalities) across Eastern, Central, and Western China. A total of 94.6% (122/129) of respondents' hospitals had adopted a telemedicine system; however, 34.1% (44/129) of the physicians had never used a telemedicine system and only 9.3% (12/129) used one frequently (≥1 time/week). A total of 91.5% (118/129) and 88.4% (114/129) of physicians were willing to use telemedicine during and after the COVID-19 pandemic, respectively. Physicians considered the inability to examine patients in person to be the biggest concern (101/129, 78.3%) and the biggest barrier (76/129, 58.9%) to implementing telemedicine. CONCLUSIONS: Telemedicine is not yet universally available for all health care needs and has not been used frequently by physicians in this study. However, the willingness of physicians to use telemedicine was high. Telemedicine still has many problems to overcome.

Understanding Behavioral Intentions Toward COVID-19 Vaccines: Theory-Based Content Analysis of Tweets.

BACKGROUND: Acceptance rates of COVID-19 vaccines have still not reached the required threshold to achieve herd immunity. Understanding why some people are willing to be vaccinated and others are not is a critical step to develop efficient implementation strategies to promote COVID-19 vaccines. OBJECTIVE: We conducted a theory-based content analysis based on the capability, opportunity, motivation-behavior (COM-B) model to characterize the factors influencing behavioral intentions toward COVID-19 vaccines mentioned on the Twitter platform. METHODS: We collected tweets posted in English from November 1-22, 2020, using a combination of relevant keywords and hashtags. After excluding retweets, we randomly selected 5000 tweets for manual coding and content analysis. We performed a content analysis informed by the adapted COM-B model. RESULTS: Of the 5000 COVID-19 vaccine-related tweets that were coded, 4796 (95.9%) were posted by unique users. A total of 97 tweets carried positive behavioral intent, while 182 tweets contained negative behavioral intent. Of these, 28 tweets were mapped to capability factors, 155 tweets were related to motivation, 23 tweets were related to opportunities, and 74 tweets did not contain any useful information about the reasons for their behavioral intentions (κ=0.73). Some tweets mentioned two or more constructs at the same time. Tweets that were mapped to capability (P<.001), motivation (P<.001), and opportunity (P=.03) factors were more likely to indicate negative behavioral intentions. CONCLUSIONS: Most behavioral intentions regarding COVID-19 vaccines were related to the motivation construct. The themes identified in this study could be used to inform theory-based and evidence-based interventions to improve acceptance of COVID-19 vaccines.

The management of coronavirus disease 2019 (COVID-19).

In December 2019, a novel coronavirus causing severe acute respiratory disease occurred in Wuhan, China. It is an emerging infectious disease with widespread and rapid infectiousness. The World Health Organization declared the coronavirus outbreak to be a public health emergency of international concern on 31 January 2020. Severe COVID-19 patients should be managed and treated in a critical care unit. Performing a chest X-ray/CT can judge the severity of the disease. The management of COVID-19 patients includes epidemiological risk and patient isolation; treatment entails general supportive care, respiratory support, symptomatic treatment, nutritional support, psychological intervention, etc. The prognosis of the patients depends upon the severity of the disease, the patient's age, the underlying diseases of the patients, and the patient's overall medical condition. The management of COVID-19 should focus on early diagnosis, immediate isolation, general and optimized supportive care, and infection prevention and control.

Epidemiology, clinical characteristics of the first cases of COVID-19.

BACKGROUND: COVID-19 is currently the most urgent threat to public health in the world. The aim of this study is to provide an overview of the first cases of COVID-19 to make further improvements in health policies and prevention measurements in response to the outbreak of COVID-19. METHODS: We performed a search in PubMed, the CNKI (China National Knowledge Infrastructure), Web of Science and the WHO database of publications on COVID-19 for peer-reviewed papers from 1 December 2019 to 9 July 2020. We analysed the demographics, epidemiological characteristics, clinical features, signs and symptoms of the disease at the onset. RESULTS: We identified the first cases of COVID-19 in 16 different countries/regions from Asia, Europe, North America and South America. Of these 16 cases, 8 (50.0%) were male, with a mean of age 43.38 ± 15.19 years. All the cases had a history of travel or exposure. Twelve cases (75.0%) occurred in January, eight patients were Chinese, two patients were international students in Wuhan, one patient had a history of travelling in Wuhan, and one patient was in contact with Chinese patient. The longest hospital stay was 24 days (1 patient), and the shortest was 5 days (1 patient). The usual hospital stay was 9 days (4 patients). CONCLUSION: Understanding the epidemiological characteristics, clinical characteristics, and diagnosis and treatment of the first patients in various countries are of great significance for the identification, prevention and control of COVID-19.