Dataset Information

Development and Validation of a Multimodal-Based Prognosis and Intervention Prediction Model for COVID-19 Patients in a Multicenter Cohort.

ABSTRACT: The ability to accurately predict the prognosis and intervention requirements for treating highly infectious diseases, such as COVID-19, can greatly support the effective management of patients, especially in resource-limited settings. The aim of the study is to develop and validate a multimodal artificial intelligence (AI) system using clinical findings, laboratory data and AI-interpreted features of chest X-rays (CXRs), and to predict the prognosis and the required interventions for patients diagnosed with COVID-19, using multi-center data. In total, 2282 real-time reverse transcriptase polymerase chain reaction-confirmed COVID-19 patients' initial clinical findings, laboratory data and CXRs were retrospectively collected from 13 medical centers in South Korea, between January 2020 and June 2021. The prognostic outcomes collected included intensive care unit (ICU) admission and in-hospital mortality. Intervention outcomes included the use of oxygen (O₂) supplementation, mechanical ventilation and extracorporeal membrane oxygenation (ECMO). A deep learning algorithm detecting 10 common CXR abnormalities (DLAD-10) was used to infer the initial CXR taken. A random forest model with a quantile classifier was used to predict the prognostic and intervention outcomes, using multimodal data. The area under the receiver operating curve (AUROC) values for the single-modal model, using clinical findings, laboratory data and the outputs from DLAD-10, were 0.742 (95% confidence interval [CI], 0.696-0.788), 0.794 (0.745-0.843) and 0.770 (0.724-0.815), respectively. The AUROC of the combined model, using clinical findings, laboratory data and DLAD-10 outputs, was significantly higher at 0.854 (0.820-0.889) than that of all other models (p < 0.001, using DeLong's test). In the order of importance, age, dyspnea, consolidation and fever were significant clinical variables for prediction. The most predictive DLAD-10 output was consolidation. We have shown that a multimodal AI model can improve the performance of predicting both the prognosis and intervention in COVID-19 patients, and this could assist in effective treatment and subsequent resource management. Further, image feature extraction using an established AI engine with well-defined clinical outputs, and combining them with different modes of clinical data, could be a useful way of creating an understandable multimodal prediction model.

SUBMITTER: Lee JH

PROVIDER: S-EPMC9269794 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundThe current COVID-19 pandemic is unprecedented; under resource-constrained settings, predictive algorithms can help to stratify disease severity, alerting physicians of high-risk patients; however, there are only few risk scores derived from a substantially large electronic health record (EHR) data set, using simplified predictors as input.ObjectiveThe objectives of this study were to develop and validate simplified machine learning algorithms that predict COVID-19 adverse outcomes; to evaluate the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and calibration of the algorithms; and to derive clinically meaningful thresholds.MethodsWe performed machine learning model development and validation via a cohort study using multicenter, patient-level, longitudinal EHRs from the Optum COVID-19 database that provides anonymized, longitudinal EHR from across the United States. The models were developed based on clinical characteristics to predict 28-day in-hospital mortality, intensive care unit (ICU) admission, respiratory failure, and mechanical ventilator usages at inpatient setting. Data from patients who were admitted from February 1, 2020, to September 7, 2020, were randomly sampled into development, validation, and test data sets; data collected from September 7, 2020, to November 15, 2020, were reserved as the postdevelopment prospective test data set.ResultsOf the 3.7 million patients in the analysis, 585,867 patients were diagnosed or tested positive for SARS-CoV-2, and 50,703 adult patients were hospitalized with COVID-19 between February 1 and November 15, 2020. Among the study cohort (n=50,703), there were 6204 deaths, 9564 ICU admissions, 6478 mechanically ventilated or EMCO patients, and 25,169 patients developed acute respiratory distress syndrome or respiratory failure within 28 days since hospital admission. The algorithms demonstrated high accuracy (AUC 0.89, 95% CI 0.89-0.89 on the test data set [n=10,752]), consistent prediction through the second wave of the pandemic from September to November (AUC 0.85, 95% CI 0.85-0.86) on the postdevelopment prospective test data set [n=14,863], great clinical relevance, and utility. Besides, a comprehensive set of 386 input covariates from baseline or at admission were included in the analysis; the end-to-end pipeline automates feature selection and model development. The parsimonious model with only 10 input predictors produced comparably accurate predictions; these 10 predictors (age, blood urea nitrogen, SpO2, systolic and diastolic blood pressures, respiration rate, pulse, temperature, albumin, and major cognitive disorder excluding stroke) are commonly measured and concordant with recognized risk factors for COVID-19.ConclusionsThe systematic approach and rigorous validation demonstrate consistent model performance to predict even beyond the period of data collection, with satisfactory discriminatory power and great clinical utility. Overall, the study offers an accurate, validated, and reliable prediction model based on only 10 clinical features as a prognostic tool to stratifying patients with COVID-19 into intermediate-, high-, and very high-risk groups. This simple predictive tool is shared with a wider health care community, to enable service as an early warning system to alert physicians of possible high-risk patients, or as a resource triaging tool to optimize health care resources.

Project description:BackgroundCoronavirus disease 2019 (COVID-19) is an infectious disease spreading rapidly worldwide. As it quickly spreads and can cause severe disease, early detection and treatment may reduce mortality. Therefore, the study aims to construct a risk model and a nomogram for predicting the mortality of COVID-19.MethodsThe original data of this study were from the article "Neurologic Syndromes Predict Higher In-Hospital Mortality in COVID-19." The database contained 4,711 multiethnic patients. In this secondary analysis, a statistical difference test was conducted for clinical demographics, clinical characteristics, and laboratory indexes. The least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression analysis were applied to determine the independent predictors for the mortality of COVID-19. A nomogram was conducted and validated according to the independent predictors. The area under the curve (AUC), the calibration curve, and the decision curve analysis (DCA) were carried out to evaluate the nomogram.ResultsThe mortality of COVID-19 is 24.4%. LASSO and multivariate logistic regression analysis suggested that risk factors for age, PCT, glucose, D-dimer, CRP, troponin, BUN, LOS, MAP, AST, temperature, O2Sats, platelets, Asian, and stroke were independent predictors of CTO. Using these independent predictors, a nomogram was constructed with good discrimination (0.860 in the C index) and internal validation (0.8479 in the C index), respectively. The calibration curves and the DCA showed a high degree of reliability and precision for this clinical prediction model.ConclusionAn early warning model based on accessible variates from routine clinical tests to predict the mortality of COVID-19 were conducted. This nomogram can be conveniently used to facilitate identifying patients who might develop severe disease at an early stage of COVID-19. Further studies are warranted to validate the prognostic ability of the nomogram.

Project description:BackgroundThere are rare prediction models for esophageal squamous cell carcinoma (ESCC) for rural Chinese population. We aimed to develop and validate a prediction model for ESCC based on a cohort study for the population.MethodsData of 115,686 participants were collected from esophageal cancer (EC) early diagnosis and treatment of cancer program as derivation cohort while data of 54,750 participants were collected as validation cohort. Risk factors considered included age, sex, smoking status, alcohol drinking status, body mass index (BMI), tea drinking status, marital status, annual household income, source of drinking water, education level, and diet habit. Cox proportional hazards model was used to develop ESCC prediction model at 5 years. Calibration ability, discrimination ability, and decision curve analysis were analyzed in both derivation and validation cohort. A score model was developed based on prediction model.ResultsOne hundred eighty-six cases were diagnosed during 556,949.40 person-years follow-up in the derivation cohort while 120 cases from 277,302.70 in the validation cohort. Prediction model included the following variables: age, sex, alcohol drinking status, BMI, tea drinking status, and fresh fruit. The model had good discrimination and calibration performance: R 2, D statistic, and Harrell's C statistic of prediction model were 43.56%, 1.70, and 0.798 in derivation cohort and 45.19%, 1.62, and 0.787 in validation cohort. The calibration analysis showed good coherence between predicted probabilities and observed probabilities while decision curve analysis showed clinical usefulness. The score model was as follows: age (3 for 45-49 years old; 4 for 50-54 years old; 7 for 55-59 years old; 9 for 60-64 years; 10 for 65-69 years), sex (5 for men), BMI (1 for ≤25), alcohol drinking status (2 for alcohol drinkers), tea drinking status (2 for tea drinkers), and fresh fruit (2 for never) and showed good discrimination ability with area under the curve and its 95% confidence interval of 0.792 (0.761,0.822) in the deviation cohort and 0.773 (0.736,0.811) in the validation cohort. The calibration analysis showed great coherence between predicted probabilities and observed probabilities.ConclusionsWe developed and validated an ESCC prediction model using cohort study with good discrimination and calibration capability which can be used for EC screening for rural Chinese population.

Dataset Information

Development and Validation of a Multimodal-Based Prognosis and Intervention Prediction Model for COVID-19 Patients in a Multicenter Cohort.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets