Machine Learning to Calculate Heparin Dose in COVID-19 Patients with Active Cancer.

Egidio Imbalzano,Giovanni Squadrito,Giuseppe Nato,Pierpaolo Di Micco,Angela Sciacqua,Luana Orlando,Marco Vatrano,Francesco Dentali,Vincenzo Russo,Arrigo F G Cicero,Gianluca Bagnato,Giuseppe Camporese,Giuseppe Dattilo,Antonio Giovanni Versace,Veronica Nassisi

doi:10.3390/jcm11010219

Abstract

To realize a machine learning (ML) model to estimate the dose of low molecular weight heparin to be administered, preventing thromboembolism events in COVID-19 patients with active cancer. Methods: We used a dataset comprising 131 patients with active cancer and COVID-19. We considered five ML models: logistic regression, decision tree, random forest, support vector machine and Gaussian naive Bayes. We decided to implement the logistic regression model for our study. A model with 19 variables was analyzed. Data were randomly split into training (70%) and testing (30%) sets. Model performance was assessed by confusion matrix metrics on the testing data for each model as positive predictive value, sensitivity and F1-score. Results: We showed that the five selected models outperformed classical statistical methods of predictive validity and logistic regression was the most effective, being able to classify with an accuracy of 81%. The most relevant result was finding a patient-proof where python function was able to obtain the exact dose of low weight molecular heparin to be administered and thereby to prevent the occurrence of VTE. Conclusions: The world of machine learning and artificial intelligence is constantly developing. The identification of a specific LMWH dose for preventing VTE in very high-risk populations, such as the COVID-19 and active cancer population, might improve with the use of new training ML-based algorithms. Larger studies are needed to confirm our exploratory results.

Highlights

COVID-19 is an acute, systemic complex disorder induced by SARS-CoV-2 infection, with heterogeneous manifestations ranging from paucisymptomatic course to lifethreatening severe presentation characterized by bilateral interstitial pneumonia and acute respiratory distress syndrome [1]
We employed an approach based on machine learning (ML), a branch of computer science that can be considered a close relative of artificial intelligence, to achieve, through an algorithm, the correct anticoagulant therapy to be administered in primary prevention to COVID-19 patients with active cancer
We excluded from the study patients who did not require low molecular weight heparin prophylaxis or who were already being treated with VKA/DOACs, and patients with a known diagnosis of pulmonary embolism or venous thrombosis

Summary

Introduction

COVID-19 is an acute, systemic complex disorder induced by SARS-CoV-2 infection, with heterogeneous manifestations ranging from paucisymptomatic course to lifethreatening severe presentation characterized by bilateral interstitial pneumonia and acute respiratory distress syndrome [1] It has been associated with a hypercoagulable state and thrombotic complications, mainly in its critical form [2]. Robin Park et al [14] in a meta-analysis of 16 retrospective and prospective studies, with 3558 patients, show an increased mortality in patients under active chemotherapy treatment, compared to not active chemotherapy For this reason, a correct evaluation of antithrombotic therapy is essential in oncologic patients, and able to reduce mortality, especially when the appropriate dosage of low molecular weight heparin (LMWH) is administered [15,16]. Machine learning techniques, compared with traditional statistical models, have many advantages including high power and accuracy, the ability to model non-linear effects, the interpretation of large genomic data sets, robustness to parameter assumptions, and the ability to dispense with a normal distribution test [17]

Objectives

Methods

Results

Discussion

Conclusion