A Machine Learning Model for the Prediction of COVID-19 Severity Using RNA-Seq, Clinical, and Co-Morbidity Data

Diagnostics, Год журнала: 2024, Номер 14(12), С. 1284 - 1284

Опубликована: Июнь 18, 2024

The premise for this study emanated from the need to understand SARS-CoV-2 infections at molecular level and develop predictive tools managing COVID-19 severity. With varied clinical outcomes observed among infected individuals, creating a reliable machine learning (ML) model predicting severity of became paramount. Despite availability large-scale genomic data, previous studies have not effectively utilized multi-modality data disease prediction using data-driven approaches. Our primary goal is predict machine-learning trained on combination patients’ gene expression, features, co-morbidity data. Employing various ML algorithms, including Logistic Regression (LR), XGBoost (XG), Naïve Bayes (NB), Support Vector Machine (SVM), alongside feature selection methods, we sought identify best-performing prediction. results highlighted XG as superior classifier, with 95% accuracy 0.99 AUC (Area Under Curve), distinguishing groups. Additionally, SHAP analysis revealed vital features contributing prediction, several genes such COX14, LAMB2, DOLK, SDCBP2, RHBDL1, IER3-AS1. Notably, two absolute neutrophil count Viremia Categories, emerged top contributors. Integrating multiple modalities has significantly improved compared any single modality. identified could serve biomarkers prognosis patient care, allowing clinicians optimize treatment strategies refine decision-making processes enhanced outcomes.

Язык: Английский

Rule-Based Cardiovascular Disease Diagnosis

Springer eBooks, Год журнала: 2023, Номер unknown, С. 740 - 750

Опубликована: Янв. 1, 2023

Язык: Английский

An Edge Based Framework for Risk Assessment of Communicable Disease

2021 13th International Conference on Wireless Communications and Signal Processing (WCSP), Год журнала: 2022, Номер 22, С. 331 - 335

Опубликована: Ноя. 1, 2022

Along with the development of edge computing and Artificial Intelligence (AI), there has been an explosion health-care system. As COVID-19 spread globally, pandemic created significant challenges for global health Therefore, we proposed edge-based framework risk assessment communicable disease called CDM-FL. The CDM-FL consists two modules, common data model (CDM) federated learning (FL). CDM can process store multi-source heterogeneous standardized semantics schema. This provides more training using medical globally. is deployed on nodes that measure patients' status locally low latency. It also keeps patient privacy from being disclosed are likely to share their data. results based real-world show help physicians evaluate as well save lives during severe epidemic situations.

Язык: Английский