COVID-19 diagnosis by routine blood tests using machine learning

Matjaž Kukar,Sašo Moškon,Peter Černelč,Miran Brvar,Marko Notar,Tomaž Vovko,Mateja Zalaznik,Mateja Notar,Gregor Gunčar,Simon Podnar

doi:10.1038/s41598-021-90265-9

Matjaž Kukar, Sašo Moškon + Show 8 more

Open Access

https://doi.org/10.1038/s41598-021-90265-9

Copy DOI

Abstract

Physicians taking care of patients with COVID-19 have described different changes in routine blood parameters. However, these changes hinder them from performing COVID-19 diagnoses. We constructed a machine learning model for COVID-19 diagnosis that was based and cross-validated on the routine blood tests of 5333 patients with various bacterial and viral infections, and 160 COVID-19-positive patients. We selected the operational ROC point at a sensitivity of 81.9% and a specificity of 97.9%. The cross-validated AUC was 0.97. The five most useful routine blood parameters for COVID-19 diagnosis according to the feature importance scoring of the XGBoost algorithm were: MCHC, eosinophil count, albumin, INR, and prothrombin activity percentage. t-SNE visualization showed that the blood parameters of the patients with a severe COVID-19 course are more like the parameters of a bacterial than a viral infection. The reported diagnostic accuracy is at least comparable and probably complementary to RT-PCR and chest CT studies. Patients with fever, cough, myalgia, and other symptoms can now have initial routine blood tests assessed by our diagnostic tool. All patients with a positive COVID-19 prediction would then undergo standard RT-PCR studies to confirm the diagnosis. We believe that our results represent a significant contribution to improvements in COVID-19 diagnosis.

Highlights

Physicians taking care of patients with COVID-19 have described different changes in routine blood parameters
When the COVID-19-positive group was compared to other viral infections, the five parameters with the statistically most significant difference and effect size were mean corpuscular hemoglobin concentration (MCHC), eosinophils ratio, prothrombin international normalized ratio (INR), prothrombin activity %, and creatinine (Fig. 2)
We confirmed that COVID-19 diagnosis is attainable using machine learning (ML) on data from routine blood tests

Summary

Introduction

Physicians taking care of patients with COVID-19 have described different changes in routine blood parameters. These changes hinder them from performing COVID-19 diagnoses. We constructed a machine learning model for COVID-19 diagnosis that was based and cross-validated on the routine blood tests of 5333 patients with various bacterial and viral infections, and 160 COVID-19positive patients. All patients with a positive COVID-19 prediction would undergo standard RT-PCR studies to confirm the diagnosis. Physicians taking care of COVID-19 patients have noted pronounced changes in their blood parameters. They have described hypoalbuminemia, increased C-reactive protein (CRP) and lactate dehydrogenase (LDH), lymphopenia, etc.[17]. Have previously demonstrated how an ML model considerably outperformed experienced clinicians in diagnosing hematological d isorders[18], as well as another model for brain tumors with diagnostic accuracy similar to head imaging[19]

Objectives

Methods

Results

Conclusion