Infoveillance study on the dynamic associations between CDC social media contents and epidemic measures during COVID-19 (preprint)

Background: Health agencies have been widely adopting social media to disseminate important information, educate the public on emerging health issues, and understand public opinions. The Centers for Disease Control and Prevention (CDC) has been one of the leading agencies that utilizes social media platforms during the COVID-19 pandemic to communicate with the public and mitigate the disease in the United States. It is crucial to understand the relationships between CDC's social media communication and the actual epidemic metrics to improve public health agencies' communication strategies during health emergencies. Objective: The aim of this study was to identify key topics in tweets posted by CDC during the pandemic, to investigate the temporal dynamics between these key topics and the actual COVID-19 epidemic measures, and to make recommendations for CDC's digital health communication strategies for future health emergencies. Methods: Two types of data were collected: 1) a total of 17,524 COVID-19-related English tweets posted by the CDC between December 7, 2019 and January 15, 2022; 2) COVID-19 epidemic measures in the U.S. from the public GitHub repository of Johns Hopkins University from January 2020 to July 2022. Latent Dirichlet allocation (LDA) topic modeling was applied to identify key topics from all COVID-19-related tweets posted by CDC, and the final topics were determined by domain experts. Various multivariate time series analysis techniques were applied between each of the identified key topics and actual COVID-19 epidemic measures to quantify the dynamic associations between these two types of time series data. Results: Four major topics from CDC's COVID-19 tweets were identified: 1) information on prevention of health outcomes of COVID-19; 2) pediatric intervention and family safety; 3) updates of the epidemic situation of COVID-19; 4) research and community engagement to curb COVID-19. Multivariate analyses showed that there were significant variabilities of progression between CDC's topics and the actual COVID-19 epidemic measures. Some CDC's topics showed substantial associations with the COVID-19 measures over different time spans throughout the pandemic, expressing similar temporal dynamics between these two types of time series data. Conclusions: Our study is the first to comprehensively investigate the dynamic associations between topics discussed by CDC on Twitter and the COVID-19 epidemic measures in the U.S. We identified four major topic themes via topic modeling and explored how each of these topics was associated with each major epidemic measure by performing various multivariate time series analyses. We recommend that it is critical for public health agencies, such as CDC, to disseminate and update timely and accurate information to the public and align major topics with the key epidemic measures over time. We suggest that social media can help public health agencies to inform the public on health emergencies and to mitigate them effectively.

An Interpretable Machine Learning Framework for Accurate Severe vs Non-Severe COVID-19 Clinical Type Classification (preprint)

Effectively and efficiently diagnosing COVID-19 patients with accurate clinical type is essential to achieve optimal outcomes for the patients as well as reducing the risk of overloading the healthcare system. Currently, severe and non-severe COVID-19 types are differentiated by only a few features, which do not comprehensively characterize the complicated pathological, physiological, and immunological responses to SARS-CoV-2 invasion in different types. In this study, we recruited 214 confirmed COVID-19 patients in non-severe and 148 in severe type, from Wuhan, China. The patients’ comorbidity and symptoms (including 26 features), and laboratory testing results (26 features) upon admission were acquired as two input modalities. Exploratory analyses demonstrated that these features differed substantially between two clinical types. Machine learning random forest (RF) models based on features in each modality were developed and validated to classify COVID-19 clinical types. Using comorbidity/symptom and laboratory results as input independently, RF models achieved >90% and >95% predictive accuracy, respectively. Input features’ importance based on Gini impurity were further evaluated and top five features from each modality were identified (age, hypertension, cardiovascular disease, gender, diabetes; D-Dimer, hsTNI, absolute neutrophil count, IL-6, and LDH, in descending order). Combining top 10 multimodal features, RF model achieved >99% predictive accuracy. These findings shed light on how the human body reacts to SARS-CoV-2 invasion as a unity and provide insights on effectively evaluating COVID-19 patient’s severity and developing personalized treatment plans accordingly. We suggest that symptoms and comorbidities can be used as an initial screening tool for triaging, while laboratory results are applied when accuracy is the priority.Funding Statement: This study was jointly supported by the National Science Foundation for Young Scientists of China (81703201), the Natural Science Foundation for Young Scientists of Jiangsu Province (BK20171076), the Jiangsu Provincial Medical Innovation Team (CXTDA2017029), the Jiangsu Provincial Medical Youth Talent program (QNRC2016548), the Jiangsu Preventive Medicine Association program (Y2018086), the Lifting Program of Jiangsu Provincial Scientific and Technological Association, and the Jiangsu Government Scholarship for Overseas Studies.Declaration of Interests: The authors declare no competing interests in this study.Ethics Approval Statement: Patient-specific identifying information (e.g., name, address of residence) was removed from data collected for this study. This study was evaluated and approved by the IRB committee of Union Hospital, Wuhan, China (approval number: 2020-IEC-J-345).

Accurately Differentiating COVID-19, Other Viral Infection, and Healthy Individuals Using Multimodal Features via Late Fusion Learning (preprint)

Effectively identifying COVID-19 patients using non-PCR clinical data is critical for the optimal clinical outcomes. Currently, there is a lack of comprehensive understanding of various biomedical features and appropriate technical approaches to accurately detecting COVID-19 patients. In this study, we recruited 214 confirmed COVID-19 patients in non-severe (NS) and 148 in severe (S) clinical type, 198 non-infected healthy (H) participants and 129 non-COVID viral pneumonia (V) patients. The participants' clinical information (23 features), lab testing results (10 features), and thoracic CT scans upon admission were acquired as three input feature modalities. To enable late fusion of multimodality data, we developed a deep learning model to extract a 10-feature high-level representation of the CT scans. Exploratory analyses showed substantial differences of all features among the four classes. Three machine learning models (k-nearest neighbor kNN, random forest RF, and support vector machine SVM) were developed based on the 43 features combined from all three modalities to differentiate four classes (NS, S, V, and H) at once. All three models had high accuracy to differentiate the overall four classes (95.4%-97.7%) and each individual class (90.6%-99.9%). Multimodal features provided substantial performance gain from using any single feature modality. Compared to existing binary classification benchmarks often focusing on single feature modality, this study provided a novel and effective breakthrough for clinical applications. Findings and the analytical workflow can be used as clinical decision support for current COVID-19 and other clinical applications with high-dimensional multimodal biomedical features.

An Interpretable Machine Learning Framework for Accurate Severe vs Non-severe COVID-19 Clinical Type Classification (preprint)

Effectively and efficiently diagnosing COVID-19 patients with accurate clinical type is essential to achieve optimal outcomes of the patients as well as reducing the risk of overloading the healthcare system. Currently, severe and non-severe COVID-19 types are differentiated by only a few clinical features, which do not comprehensively characterize complicated pathological, physiological, and immunological responses to SARS-CoV-2 invasion in different types. In this study, we recruited 214 confirmed COVID-19 patients in non-severe and 148 in severe type, from Wuhan, China. The patients' comorbidity and symptoms (26 features), and blood biochemistry (26 features) upon admission were acquired as two input modalities. Exploratory analyses demonstrated that these features differed substantially between two clinical types. Machine learning random forest (RF) models using features in each modality were developed and validated to classify COVID-19 clinical types. Using comorbidity/symptom and biochemistry as input independently, RF models achieved >90% and >95% predictive accuracy, respectively. Input features' importance based on Gini impurity were further evaluated and top five features from each modality were identified (age, hypertension, cardiovascular disease, gender, diabetes; D-Dimer, hsTNI, neutrophil, IL-6, and LDH). Combining top 10 multimodal features, RF model achieved >99% predictive accuracy. These findings shed light on how the human body reacts to SARS-CoV-2 invasion as a unity and provide insights on effectively evaluating COVID-19 patient's severity and developing treatment plans accordingly. We suggest that symptoms and comorbidities can be used as an initial screening tool for triaging, while biochemistry and features combined are applied when accuracy is the priority.

Triaging moderate COVID-19 and other viral pneumonias from routine blood tests (preprint)

The COVID-19 is sweeping the world with deadly consequences. Its contagious nature and clinical similarity to other pneumonias make separating subjects contracted with COVID-19 and non-COVID-19 viral pneumonia a priority and a challenge. However, COVID-19 testing has been greatly limited by the availability and cost of existing methods, even in developed countries like the US. Intrigued by the wide availability of routine blood tests, we propose to leverage them for COVID-19 testing using the power of machine learning. Two proven-robust machine learning model families, random forests (RFs) and support vector machines (SVMs), are employed to tackle the challenge. Trained on blood data from 208 moderate COVID-19 subjects and 86 subjects with non-COVID-19 moderate viral pneumonia, the best result is obtained in an SVM-based classifier with an accuracy of 84%, a sensitivity of 88%, a specificity of 80%, and a precision of 92%. The results are found explainable from both machine learning and medical perspectives. A privacy-protected web portal is set up to help medical personnel in their practice and the trained models are released for developers to further build other applications. We hope our results can help the world fight this pandemic and welcome clinical verification of our approach on larger populations.