Problem / Objective

The COVID-19 pandemic has put immense pressure on the healthcare systems around the globe. In these times, it is crucial to assess risk so that critical resources can be mobilised to treat patients progressing to severe stages. Focused medical treatments can be administered only when there is a clear understanding of the risk factors that influence mortality the most. Therefore, it has become imperative to know early on in the diagnosis about the progression of the disease. Machine learning methods are capable of discerning useful patterns in large dimensional data that is expected to aid in the decision-making process of identifying, with high accuracy, patients who are at high risk. 

Solution / Approach

Researchers from Centre for Computational Natural Sciences and Bioinformatics (CCNSB) at IIIT-H have developed an ML model for risk and mortality prediction of COVID-19 patients. A powerful combination of five features: age, neutrophils, lymphocytes, LDH, and hs-CRP, has helped to predict mortality with 96% accuracy. In their study titled ‘Machine learning based clinical decision support system for early COVID-19 mortality prediction’, the researchers have attempted to provide a mortality prediction as early as 16 days before the outcome. The outcome in this case is described as discharge from hospital or death. Various ML models (neural networks, logistic regression, XGBoost, random forests, SVM, and decision trees) have been trained and performance compared to determine the model that achieves consistently high accuracy across the days that span the disease. The best performing method using XGBoost feature importance and neural network classification, predicts with an accuracy of 90% as early as 16 days before the outcome. Robust testing with three cases based on days to outcome confirms the strong predictive performance and practicality of the proposed model. A detailed analysis and identification of trends was performed using these key biomarkers to provide useful insights for intuitive application.

Impact /Implementation

This study provide solutions that would help accelerate the decision-making process in healthcare systems for focused medical treatments in an accurate, early, and reliable manner.

Image by Tumisu from Pixabay