AsianScientist (May 14, 2020) – A machine learning model has selected three biomarkers that can predict the mortality of COVID-19 patients with more than 90 percent accuracy, according to paper published today in Nature Machine Intelligence.
Fast, accurate and early clinical assessment of patients’ COVID-19 severity is vital. However, there is no useful predictive biomarker to distinguish patients that require immediate medical attention and to estimate their associated mortality rate.
Yuan Ye, Xu Hui and Li Shusheng from the Huazhong University of Science and Technology in Wuhan, China, retrospectively analyzed the blood samples of 485 patients from Wuhan city to identify robust and meaningful markers of mortality risk.
To develop their machine-learning model, they collected blood samples between January 10 and February 18 from 485 COVID-19 patients in Tongji Hospital. Of the 375 cases included in the analysis, 201 recovered from COVID-19 and were discharged from the hospital, while the remaining 174 patients died.
The authors designed a mathematical modeling approach based on machine learning algorithms devised to identify the biomarkers most predictive of patient mortality. The problem was formulated as a classification task, where the inputs included basic information, symptoms, blood samples and the results of laboratory tests from a group that included general, severe and critical patients.
“Through optimization, this classifier aims to reveal the most crucial biomarkers distinguishing patients at imminent risk, thereby relieving clinical burden and potentially reducing the mortality rate,” the authors write in the paper.
The model selected lactic dehydrogenase (LDH), lymphocyte and high-sensitivity C-reactive protein (hs-CRP) levels as the most crucial biomarkers distinguishing patients at imminent risk. In particular, relatively high levels of LDH alone seem to play a crucial role in distinguishing the vast majority of cases that require immediate medical attention. This finding is consistent with current medical knowledge that high LDH levels are associated with tissue breakdown occurring in various diseases, including pulmonary disorders such as pneumonia.
“The significance of our work is twofold. First, it goes beyond providing high-risk factors. It provides a simple and intuitive clinical test to precisely and quickly quantify the risk of death,” the authors write, adding that this procedure will need to be repeated for better accuracy as more data become available.
“Second, the three key features, LDH, lymphocytes and hs-CRP, can be easily collected in any hospital. In crowded hospitals, and with shortages of medical resources, this simple model can help to quickly prioritize patients, especially during a pandemic when limited healthcare resources have to be allocated.”
The article can be found at: Yan et al. (2020) An Interpretable Mortality Prediction Model for COVID-19 Patients.
———
Source: Huazhong University of Science and Technology; Photo: Shutterstock.
Disclaimer: This article does not necessarily reflect the views of AsianScientist or its staff.
