Multivariable Risk Modelling and Survival Analysis with Machine Learning in SARS-CoV-2 Infection

Abstract

Aim: To evaluate the performance of a machine learning model based on demographic variables, blood tests, pre-existing comorbidities, and computed tomography(CT)-based radiomic features to predict critical outcome in patients with acute respiratory syndrome coronavirus 2 (SARS-CoV-2).

Methods: We retrospectively enrolled 694 SARS-CoV-2-positive patients. Clinical and demographic data were extracted from clinical records. Radiomic data were extracted from CT. Patients were randomized to the training (80%, n = 556) or test (20%, n = 138) dataset. The training set was used to define the association between severity of disease and comorbidities, laboratory tests, demographic, and CT-based radiomic variables, and to implement a risk-prediction model. The model was evaluated using the C statistic and Brier scores. The test set was used to assess model prediction performance.

Results: Patients who died (n = 157) were predominantly male (66%) over the age of 50 with median (range) C-reactive protein (CRP) = 5 [1, 37] mg/dL, lactate dehydrogenase (LDH) = 494 [141, 3631] U/I, and D-dimer = 6.006 [168, 152.015] ng/mL. Surviving patients (n = 537) had median (range) CRP = 3 [0, 27] mg/dL, LDH = 484 [78, 3.745] U/I, and D-dimer = 1.133 [96, 55.660] ng/mL. The strongest risk factors were D-dimer, age, and cardiovascular disease. The model implemented using the variables identified using the LASSO Cox regression analysis classified 90% of non-survivors as high-risk individuals in the testing dataset. In this sample, the estimated median survival in the high-risk group was 9 days (95% CI; 9-37), while the low-risk group did not reach the median survival of 50% (p < 0.001).

Conclusions: A machine learning model based on combined data available on the first days of hospitalization (demographics, CT-radiomics, comorbidities, and blood biomarkers), can identify SARS-CoV-2 patients at risk of serious illness and death.

Keywords: CT; SARS-CoV-2; machine learning; radiomics; survival.

Figure 1

Flowchart for machine learning model development.

Figure 2

Predictors of outcome. (A) Coefficient profile plotted versus the log (λ). Each colored line represents the coefficient of each feature. (B) The C-index was plotted versus log (λ). The green circle and line locate the Lambda with minimum cross-validation error. The blue circle and line locate the point with minimum cross-validation error plus one standard deviation. (C) Variables that survived the LASSO regression, including age, D-dimer, LDH, three comorbidities, and four radiomic variables.

Figure 3

Survival curves. Training data set (A): the survival time of SARS-CoV-2 patients in the high-risk group differed significantly from that of the low-risk subjects, with a median of 12 days (95% CI; 10–14). The low-risk group did not achieve the 50% survival rate. Test dataset (B): the median survival duration of the high-risk group was 9 days (95% CI; 6–37) and low-risk patients did not reach the median survival of 50%.