Machine-learning-derived predictive score for early estimation of COVID-19 mortality risk in hospitalized patients

Abstract

The clinical course of COVID-19 is highly variable. It is therefore essential to predict as early and accurately as possible the severity level of the disease in a COVID-19 patient who is admitted to the hospital. This means identifying the contributing factors of mortality and developing an easy-to-use score that could enable a fast assessment of the mortality risk using only information recorded at the hospitalization. A large database of adult patients with a confirmed diagnosis of COVID-19 (n = 15,628; with 2,846 deceased) admitted to Spanish hospitals between December 2019 and July 2020 was analyzed. By means of multiple machine learning algorithms, we developed models that could accurately predict their mortality. We used the information about classifiers’ performance metrics and about importance and coherence among the predictors to define a mortality score that can be easily calculated using a minimal number of mortality predictors and yielded accurate estimates of the patient severity status. The optimal predictive model encompassed five predictors (age, oxygen saturation, platelets, lactate dehydrogenase, and creatinine) and yielded a satisfactory classification of survived and deceased patients (area under the curve: 0.8454 with validation set). These five predictors were additionally used to define a mortality score for COVID-19 patients at their hospitalization. This score is not only easy to calculate but also to interpret since it ranges from zero to eight, along with a linear increase in the mortality risk from 0% to 80%. A simple risk score based on five commonly available clinical variables of adult COVID-19 patients admitted to hospital is able to accurately discriminate their mortality probability, and its interpretation is straightforward and useful.

Predictors of mortality in hospitalised patients with COVID-19: a 1-year case–control study

“…Several authors have developed similar approaches involving different variables in the prediction of mortality; some of them include dyspnoea,41 42 BUN, platelet count,42 43 sex,42 age,41 43–46 cough,41 weight,44 cardiovascular disease,41 orotracheal intubation,41 and pleural effusion,41 44 respiratory rate,42 fraction of inspired oxygen,42 blood oxygen saturation42 43 pH,42 aspartate aminotransferase levels,42 estimated glomerular filtration rate,42 lymphocyte count,44 47 WCC,41 43 creatine46 lactic acid47 and serum calcium 47. However, most of their studies did not focus on the interaction of such variables on hospital admission, and their interpretability by clinicians is difficult.…”

Extracting relevant predictive variables for COVID-19 severity prognosis: An exhaustive comparison of feature selection techniques

“…This is in agreement with [83] -who concluded that SpO 2 <90% was a strong predictor of COVID-19 in-hospital mortality-, or with [4] -who found RR to be a significant predictor for major clinical deterioration: admission in intensive care unit (ICU), or death-. Various works adhering to ML methodologies also found oxygen saturation [2,[6][7][8]11] and/or RR [2,6,7] among their selected predictors.…”

“…In addition, different authors have also addressed the performance of various ML strategies for mortality prognosis (e.g. [ 10 , 11 ]).…”

“…[ 4 ]), correlation with the target variable and importance estimates available within the trained ML algorithm (e.g. [ 6 , 11 ]). Embedded methods (such as L 1 -penalised models) were used by [ 5 , 9 ].…”

Risk-based Evaluation of ML Classification Methods Used for Medical Devices

“…In total, only 3 out of the 30 publications, i.e. the papers ( 30), (34), and (37), included a risk-based approach for the performance assessment of the ML models, in some sense. Basically, all of these three publications were addressing risk prediction as the major application.…”