1
|
Rajwa B, Naved MMA, Adibuzzaman M, Grama AY, Khan BA, Dundar MM, Rochet JC. Identification of predictive patient characteristics for assessing the probability of COVID-19 in-hospital mortality. PLOS DIGITAL HEALTH 2024; 3:e0000327. [PMID: 38652722 PMCID: PMC11037536 DOI: 10.1371/journal.pdig.0000327] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 07/15/2023] [Accepted: 03/06/2024] [Indexed: 04/25/2024]
Abstract
As the world emerges from the COVID-19 pandemic, there is an urgent need to understand patient factors that may be used to predict the occurrence of severe cases and patient mortality. Approximately 20% of SARS-CoV-2 infections lead to acute respiratory distress syndrome caused by the harmful actions of inflammatory mediators. Patients with severe COVID-19 are often afflicted with neurologic symptoms, and individuals with pre-existing neurodegenerative disease have an increased risk of severe COVID-19. Although collectively, these observations point to a bidirectional relationship between severe COVID-19 and neurologic disorders, little is known about the underlying mechanisms. Here, we analyzed the electronic health records of 471 patients with severe COVID-19 to identify clinical characteristics most predictive of mortality. Feature discovery was conducted by training a regularized logistic regression classifier that serves as a machine-learning model with an embedded feature selection capability. SHAP analysis using the trained classifier revealed that a small ensemble of readily observable clinical features, including characteristics associated with cognitive impairment, could predict in-hospital mortality with an accuracy greater than 0.85 (expressed as the area under the ROC curve of the classifier). These findings have important implications for the prioritization of clinical measures used to identify patients with COVID-19 (and, potentially, other forms of acute respiratory distress syndrome) having an elevated risk of death.
Collapse
Affiliation(s)
- Bartek Rajwa
- Bindley Bioscience Center, Purdue University, West Lafayette, Indiana, United States of America
- Purdue Institute for Integrative Neuroscience, Purdue University, West Lafayette, Indiana, United States of America
| | | | - Mohammad Adibuzzaman
- Oregon Clinical and Translational Research Institute, Oregon Health and Science University, Portland, Oregon, United States of America
| | - Ananth Y. Grama
- Dept. of Computer Science, Purdue University, West Lafayette, Indiana, United States of America
| | - Babar A. Khan
- Regenstrief Institute, Indianapolis, Indiana, United States of America
| | - M. Murat Dundar
- Dept. of Computer and Information Science, IUPUI, Indianapolis, Indiana, United States of America
| | - Jean-Christophe Rochet
- Purdue Institute for Integrative Neuroscience, Purdue University, West Lafayette, Indiana, United States of America
- Borch Dept. of Medicinal Chemistry and Molecular Pharmacology, Purdue University, West Lafayette, Indiana, United States of America
| |
Collapse
|
2
|
Wang W, Harrou F, Dairi A, Sun Y. Stacked deep learning approach for efficient SARS-CoV-2 detection in blood samples. Artif Intell Med 2024; 148:102767. [PMID: 38325923 DOI: 10.1016/j.artmed.2024.102767] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/28/2023] [Revised: 01/02/2024] [Accepted: 01/05/2024] [Indexed: 02/09/2024]
Abstract
Identifying COVID-19 through blood sample analysis is crucial in managing the disease and improving patient outcomes. Despite its advantages, the current test demands certified laboratories, expensive equipment, trained personnel, and 3-4 h for results, with a notable false-negative rate of 15%-20%. This study proposes a stacked deep-learning approach for detecting COVID-19 in blood samples to distinguish uninfected individuals from those infected with the virus. Three stacked deep learning architectures, namely the StackMean, StackMax, and StackRF algorithms, are introduced to improve the detection quality of single deep learning models. To counter the class imbalance phenomenon in the training data, the Synthetic Minority Oversampling Technique (SMOTE) algorithm is also implemented, resulting in increased specificity and sensitivity. The efficacy of the methods is assessed by utilizing blood samples obtained from hospitals in Brazil and Italy. Results revealed that the StackMax method greatly boosted the deep learning and traditional machine learning methods' capability to distinguish COVID-19-positive cases from normal cases, while SMOTE increased the specificity and sensitivity of the stacked models. Hypothesis testing is performed to determine if there is a significant statistical difference in the performance between the compared detection methods. Additionally, the significance of blood sample features in identifying COVID-19 is analyzed using the XGBoost (eXtreme Gradient Boosting) technique for feature importance identification. Overall, this methodology could potentially enhance the timely and precise identification of COVID-19 in blood samples.
Collapse
Affiliation(s)
- Wu Wang
- Center for Applied Statistics and School of Statistics, Renmin University of China, Beijing 100872, China.
| | - Fouzi Harrou
- King Abdullah University of Science and Technology (KAUST), Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division, Thuwal 23955-6900, Saudi Arabia.
| | - Abdelkader Dairi
- Computer Science Department, University of Science and Technology of Oran-Mohamed Boudiaf (USTO-MB), El Mnaouar, BP 1505, 31000, Bir El Djir, Algeria.
| | - Ying Sun
- King Abdullah University of Science and Technology (KAUST), Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division, Thuwal 23955-6900, Saudi Arabia
| |
Collapse
|
3
|
Li H, Zhao J, Chen R, Liu H, Xu X, Xu J, Jiang X, Pang M, Wang J, Li S, Hou J, Kong F. The relationships of preventive behaviors and psychological resilience with depression, anxiety, and stress among university students during the COVID-19 pandemic: A two-wave longitudinal study in Shandong Province, China. Front Public Health 2023; 11:1078744. [PMID: 37026148 PMCID: PMC10070798 DOI: 10.3389/fpubh.2023.1078744] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/24/2022] [Accepted: 02/16/2023] [Indexed: 04/08/2023] Open
Abstract
Introduction Studies have shown that the psychological impact of the COVID-19 pandemic may lead to long-term health problems; therefore, more attention should be paid to the mental health of university students. This study aimed to explore the longitudinal effects of preventive behaviors and psychological resilience on the mental health of Chinese college students during COVID-19. Methods We recruited 2,948 university students from five universities in Shandong Province. We used a generalized estimating equation (GEE) model to estimate the impact of preventive behaviors and psychological resilience on mental health. Results In the follow-up survey, the prevalence of anxiety (44.8% at T1 vs 41.2% at T2) and stress (23.0% at T1 vs 19.6% at T2) decreased over time, whereas the prevalence of depression (35.2% at T1 vs 36.9% at T2) increased significantly (P < 0.001). Senior students were more likely to report depression (OR = 1.710, P < 0.001), anxiety (OR = 0.815, P = 0.019), and stress (OR = 1.385, P = 0.011). Among all majors, medical students were most likely to report depression (OR = 1.373, P = 0.021), anxiety (OR = 1.310, P = 0.040), and stress (OR = 1.775, P < 0.001). Students who wore a mask outside were less likely to report depression (OR = 0.761, P = 0.027) and anxiety (OR = 0.686, P = 0.002) compared to those who did not wear masks. Students who complied with the standard hand-washing technique were less likely to report depression (OR = 0.628, P < 0.001), anxiety (OR = 0.701, P < 0.001), and stress (OR = 0.638, P < 0.001). Students who maintained a distance of one meter in queues were less likely to report depression (OR = 0.668, P < 0.001), anxiety (OR = 0.634, P < 0.001), and stress (OR = 0.638, P < 0.001). Psychological resilience was a protective factor against depression (OR = 0.973, P < 0.001), anxiety (OR = 0.980, P < 0.001), and stress (OR = 0.976, P < 0.001). Discussion The prevalence of depression among university students increased at follow-up, while the prevalence of anxiety and stress decreased. Senior students and medical students are vulnerable groups. University students should continue to follow relevant preventive behaviors to protect their mental health. Improving psychological resilience may help maintain and promote university students' mental health.
Collapse
Affiliation(s)
- Hexian Li
- Center for Health Management and Policy Research, School of Public Health, Cheeloo College of Medicine, Shandong University, Jinan, China
- NHC Key Lab of Health Economics and Policy Research, Shandong University, Jinan, China
| | - Jingjing Zhao
- School of Marxism, Shandong University, Jinan, China
| | - Rui Chen
- Center for Health Management and Policy Research, School of Public Health, Cheeloo College of Medicine, Shandong University, Jinan, China
- NHC Key Lab of Health Economics and Policy Research, Shandong University, Jinan, China
| | - Hui Liu
- Center for Health Management and Policy Research, School of Public Health, Cheeloo College of Medicine, Shandong University, Jinan, China
- NHC Key Lab of Health Economics and Policy Research, Shandong University, Jinan, China
| | - Xixing Xu
- Center for Health Management and Policy Research, School of Public Health, Cheeloo College of Medicine, Shandong University, Jinan, China
- NHC Key Lab of Health Economics and Policy Research, Shandong University, Jinan, China
| | - Jing Xu
- Center for Health Management and Policy Research, School of Public Health, Cheeloo College of Medicine, Shandong University, Jinan, China
- NHC Key Lab of Health Economics and Policy Research, Shandong University, Jinan, China
| | - Xiaoxu Jiang
- Center for Health Management and Policy Research, School of Public Health, Cheeloo College of Medicine, Shandong University, Jinan, China
- NHC Key Lab of Health Economics and Policy Research, Shandong University, Jinan, China
| | - Mingli Pang
- Center for Health Management and Policy Research, School of Public Health, Cheeloo College of Medicine, Shandong University, Jinan, China
- NHC Key Lab of Health Economics and Policy Research, Shandong University, Jinan, China
| | - Jieru Wang
- Center for Health Management and Policy Research, School of Public Health, Cheeloo College of Medicine, Shandong University, Jinan, China
- NHC Key Lab of Health Economics and Policy Research, Shandong University, Jinan, China
| | - Shixue Li
- Center for Health Management and Policy Research, School of Public Health, Cheeloo College of Medicine, Shandong University, Jinan, China
- NHC Key Lab of Health Economics and Policy Research, Shandong University, Jinan, China
| | - Jiaxiang Hou
- Shandong Center for Disease Control and Prevention, and Academy of Preventive Medicine, Shandong University, Jinan, China
- *Correspondence: Jiaxiang Hou
| | - Fanlei Kong
- Center for Health Management and Policy Research, School of Public Health, Cheeloo College of Medicine, Shandong University, Jinan, China
- NHC Key Lab of Health Economics and Policy Research, Shandong University, Jinan, China
- Fanlei Kong
| |
Collapse
|