Dataset Information

Prediction and Feature Importance Analysis for Severity of COVID-19 in South Korea Using Artificial Intelligence: Model Development and Validation.

ABSTRACT:

Background

The number of deaths from COVID-19 continues to surge worldwide. In particular, if a patient's condition is sufficiently severe to require invasive ventilation, it is more likely to lead to death than to recovery.

Objective

The goal of our study was to analyze the factors related to COVID-19 severity in patients and to develop an artificial intelligence (AI) model to predict the severity of COVID-19 at an early stage.

Methods

We developed an AI model that predicts severity based on data from 5601 COVID-19 patients from all national and regional hospitals across South Korea as of April 2020. The clinical severity of COVID-19 was divided into two categories: low and high severity. The condition of patients in the low-severity group corresponded to no limit of activity, oxygen support with nasal prong or facial mask, and noninvasive ventilation. The condition of patients in the high-severity group corresponded to invasive ventilation, multi-organ failure with extracorporeal membrane oxygenation required, and death. For the AI model input, we used 37 variables from the medical records, including basic patient information, a physical index, initial examination findings, clinical findings, comorbid diseases, and general blood test results at an early stage. Feature importance analysis was performed with AdaBoost, random forest, and eXtreme Gradient Boosting (XGBoost); the AI model for predicting COVID-19 severity among patients was developed with a 5-layer deep neural network (DNN) with the 20 most important features, which were selected based on ranked feature importance analysis of 37 features from the comprehensive data set. The selection procedure was performed using sensitivity, specificity, accuracy, balanced accuracy, and area under the curve (AUC).

Results

We found that age was the most important factor for predicting disease severity, followed by lymphocyte level, platelet count, and shortness of breath or dyspnea. Our proposed 5-layer DNN with the 20 most important features provided high sensitivity (90.2%), specificity (90.4%), accuracy (90.4%), balanced accuracy (90.3%), and AUC (0.96).

Conclusions

Our proposed AI model with the selected features was able to predict the severity of COVID-19 accurately. We also made a web application so that anyone can access the model. We believe that sharing the AI model with the public will be helpful in validating and improving its performance.

SUBMITTER: Chung H

PROVIDER: S-EPMC8057199 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:The SARS-CoV-2 pandemic highlighted the need for software tools that could facilitate patient triage regarding potential disease severity or even death. In this article, an ensemble of Machine Learning (ML) algorithms is evaluated in terms of predicting the severity of their condition using plasma proteomics and clinical data as input. An overview of AI-based technical developments to support COVID-19 patient management is presented outlining the landscape of relevant technical developments. Based on this review, the use of an ensemble of ML algorithms that analyze clinical and biological data (i.e., plasma proteomics) of COVID-19 patients is designed and deployed to evaluate the potential use of AI for early COVID-19 patient triage. The proposed pipeline is evaluated using three publicly available datasets for training and testing. Three ML "tasks" are defined, and several algorithms are tested through a hyperparameter tuning method to identify the highest-performance models. As overfitting is one of the typical pitfalls for such approaches (mainly due to the size of the training/validation datasets), a variety of evaluation metrics are used to mitigate this risk. In the evaluation procedure, recall scores ranged from 0.6 to 0.74 and F1-score from 0.62 to 0.75. The best performance is observed via Multi-Layer Perceptron (MLP) and Support Vector Machines (SVM) algorithms. Additionally, input data (proteomics and clinical data) were ranked based on corresponding Shapley additive explanation (SHAP) values and evaluated for their prognosticated capacity and immuno-biological credence. This "interpretable" approach revealed that our ML models could discern critical COVID-19 cases predominantly based on patient's age and plasma proteins on B cell dysfunction, hyper-activation of inflammatory pathways like Toll-like receptors, and hypo-activation of developmental and immune pathways like SCF/c-Kit signaling. Finally, the herein computational workflow is corroborated in an independent dataset and MLP superiority along with the implication of the abovementioned predictive biological pathways are corroborated. Regarding limitations of the presented ML pipeline, the datasets used in this study contain less than 1000 observations and a significant number of input features hence constituting a high-dimensional low-sample (HDLS) dataset which could be sensitive to overfitting. An advantage of the proposed pipeline is that it combines biological data (plasma proteomics) with clinical-phenotypic data. Thus, in principle, the presented approach could enable patient triage in a timely fashion if used on already trained models. However, larger datasets and further systematic validation are needed to confirm the potential clinical value of this approach. The code is available on Github: https://github.com/inab-certh/Predicting-COVID-19-severity-through-interpretable-AI-analysis-of-plasma-proteomics.

Project description:Background Cardiac MRI measurements have diagnostic and prognostic value in the evaluation of cardiopulmonary disease. Artificial intelligence approaches to automate cardiac MRI segmentation are emerging but require clinical testing. Purpose To develop and evaluate a deep learning tool for quantitative evaluation of cardiac MRI functional studies and assess its use for prognosis in patients suspected of having pulmonary hypertension. Materials and Methods A retrospective multicenter and multivendor data set was used to develop a deep learning-based cardiac MRI contouring model using a cohort of patients suspected of having cardiopulmonary disease from multiple pathologic causes. Correlation with same-day right heart catheterization (RHC) and scan-rescan repeatability was assessed in prospectively recruited participants. Prognostic impact was assessed using Cox proportional hazard regression analysis of 3487 patients from the ASPIRE (Assessing the Severity of Pulmonary Hypertension In a Pulmonary Hypertension Referral Centre) registry, including a subset of 920 patients with pulmonary arterial hypertension. The generalizability of the automatic assessment was evaluated in 40 multivendor studies from 32 centers. Results The training data set included 539 patients (mean age, 54 years ± 20 [SD]; 315 women). Automatic cardiac MRI measurements were better correlated with RHC parameters than were manual measurements, including left ventricular stroke volume (r = 0.72 vs 0.68; P = .03). Interstudy repeatability of cardiac MRI measurements was high for all automatic measurements (intraclass correlation coefficient range, 0.79-0.99) and similarly repeatable to manual measurements (all paired t test P > .05). Automated right ventricle and left ventricle cardiac MRI measurements were associated with mortality in patients suspected of having pulmonary hypertension. Conclusion An automatic cardiac MRI measurement approach was developed and tested in a large cohort of patients, including a broad spectrum of right ventricular and left ventricular conditions, with internal and external testing. Fully automatic cardiac MRI assessment correlated strongly with invasive hemodynamics, had prognostic value, were highly repeatable, and showed excellent generalizability. Clinical trial registration no. NCT03841344 Published under a CC BY 4.0 license. Online supplemental material is available for this article. See also the editorial by Ambale-Venkatesh and Lima in this issue. An earlier incorrect version appeared online. This article was corrected on June 27, 2022.

Project description:Aptamers are short single-stranded DNA, RNA, or synthetic Xeno nucleic acids (XNA) molecules that can interact with corresponding targets with high affinity. Owing to their unique features, including low cost of production, easy chemical modification, high thermal stability, reproducibility, as well as low levels of immunogenicity and toxicity, aptamers can be used as an alternative to antibodies in diagnostics and therapeutics. Systematic evolution of ligands by exponential enrichment (SELEX), an experimental approach for aptamer screening, allows the selection and identification of in vitro aptamers with high affinity and specificity. However, the SELEX process is time consuming and characterization of the representative aptamer candidates from SELEX is rather laborious. Artificial intelligence (AI) could help to rapidly identify the potential aptamer candidates from a vast number of sequences. This review discusses the advancements of AI pipelines/methods, including structure-based and machine/deep learning-based methods, for predicting the binding ability of aptamers to targets. Structure-based methods are the most used in computer-aided drug design. For this part, we review the secondary and tertiary structure prediction methods for aptamers, molecular docking, as well as molecular dynamic simulation methods for aptamer-target binding. We also performed analysis to compare the accuracy of different secondary and tertiary structure prediction methods for aptamers. On the other hand, advanced machine-/deep-learning models have witnessed successes in predicting the binding abilities between targets and ligands in drug discovery and thus potentially offer a robust and accurate approach to predict the binding between aptamers and targets. The research utilizing machine-/deep-learning techniques for prediction of aptamer-target binding is limited currently. Therefore, perspectives for models, algorithms, and implementation strategies of machine/deep learning-based methods are discussed. This review could facilitate the development and application of high-throughput and less laborious in silico methods in aptamer selection and characterization.