Dataset Information

Individual-Level Fatality Prediction of COVID-19 Patients Using AI Methods.

ABSTRACT: The global covid-19 pandemic puts great pressure on medical resources worldwide and leads healthcare professionals to question which individuals are in imminent need of care. With appropriate data of each patient, hospitals can heuristically predict whether or not a patient requires immediate care. We adopted a deep learning model to predict fatality of individuals tested positive given the patient's underlying health conditions, age, sex, and other factors. As the allocation of resources toward a vulnerable patient could mean the difference between life and death, a fatality prediction model serves as a valuable tool to healthcare workers in prioritizing resources and hospital space. The models adopted were evaluated and refined using the metrics of accuracy, specificity, and sensitivity. After data preprocessing and training, our model is able to predict whether a covid-19 confirmed patient is likely to be dead or not, given their information and disposition. The metrics between the different models are compared. Results indicate that the deep learning model outperforms other machine learning models to solve this rare event prediction problem.

SUBMITTER: Li Y

PROVIDER: S-EPMC7556112 | biostudies-literature | 2020

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Individual-Level Fatality Prediction of COVID-19 Patients Using AI Methods.

Li Yun Y Horowitz Melanie Alfonzo MA Liu Jiakang J Chew Aaron A Lan Hai H Liu Qian Q Sha Dexuan D Yang Chaowei C

Frontiers in public health 20200930

The global covid-19 pandemic puts great pressure on medical resources worldwide and leads healthcare professionals to question which individuals are in imminent need of care. With appropriate data of each patient, hospitals can heuristically predict whether or not a patient requires immediate care. We adopted a deep learning model to predict fatality of individuals tested positive given the patient's underlying health conditions, age, sex, and other factors. As the allocation of resources toward ...[more]

PMID: 33102426

Similar Datasets

Project description:Introduction: To explore the involvement of the cardiovascular system in coronavirus disease 2019 (COVID-19), we investigated whether myocardial injury occurred in COVID-19 patients and assessed the performance of serum high-sensitivity cardiac Troponin I (hs-cTnI) levels in predicting disease severity and 30-day in-hospital fatality. Methods: We included 244 COVID-19 patients, who were admitted to Renmin Hospital of Wuhan University with no preexisting cardiovascular disease or renal dysfunction. We analyzed the data including patients' clinical characteristics, cardiac biomarkers, severity of medical conditions, and 30-day in-hospital fatality. We performed multivariable Cox regressions and the receiver operating characteristic analysis to assess the association of cardiac biomarkers on admission with disease severity and prognosis. Results: In this retrospective observational study, 11% of COVID-19 patients had increased hs-cTnI levels (>40 ng/L) on admission. Of note, serum hs-cTnI levels were positively associated with the severity of medical conditions (median [interquartile range (IQR)]: 6.00 [6.00-6.00] ng/L in 91 patients with moderate conditions, 6.00 [6.00-18.00] ng/L in 107 patients with severe conditions, and 11.00 [6.00-56.75] ng/L in 46 patients with critical conditions, P for trend=0.001). Moreover, compared with those with normal cTnI levels, patients with increased hs-cTnI levels had higher in-hospital fatality (adjusted hazard ratio [95% CI]: 4.79 [1.46-15.69]). The receiver-operating characteristic curve analysis suggested that the inclusion of hs-cTnI levels into a panel of empirical prognostic factors substantially improved the prediction performance for severe or critical conditions (area under the curve (AUC): 0.71 (95% CI: 0.65-0.78) vs. 0.65 (0.58-0.72), P=0.01), as well as for 30-day fatality (AUC: 0.91 (0.85-0.96) vs. 0.77 (0.62-0.91), P=0.04). A cutoff value of 20 ng/L of hs-cTnI level led to the best prediction to 30-day fatality. Conclusions: In COVID-19 patients with no preexisting cardiovascular disease, 11% had increased hs-cTnI levels. Besides empirical prognostic factors, serum hs-cTnI levels upon admission provided independent prediction to both the severity of the medical condition and 30-day in-hospital fatality. These findings may shed important light on the clinical management of COVID-19.

Dataset Information

Individual-Level Fatality Prediction of COVID-19 Patients Using AI Methods.

Publications

Individual-Level Fatality Prediction of COVID-19 Patients Using AI Methods.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets