1. Development, evaluation, and validation of machine learning models for COVID-19 detection based on routine blood tests
- Author
-
Banfi Giuseppe, Locatelli Massimo, Campagner Andrea, Carobene Anna, Di Resta Chiara, Cabitza Federico, Sabetta Eleonora, Ferrari Davide, Ceriotti Daniele, Colombini Alessandra, De Vecchi Elena, Cabitza, Federico, Campagner, Andrea, Ferrari, Davide, Di Resta, Chiara, Ceriotti, Daniele, Sabetta, Eleonora, Colombini, Alessandra, De Vecchi, Elena, Banfi, Giuseppe, Locatelli, Massimo, Carobene, Anna, Cabitza, F, Campagner, A, Ferrari, D, Di Resta, C, Ceriotti, D, Sabetta, E, Colombini, A, De Vecchi, E, Banfi, G, Locatelli, M, and Carobene, A
- Subjects
complete blood count ,Coronavirus disease 2019 (COVID-19) ,Clinical Biochemistry ,Datasets as Topic ,Economic shortage ,030204 cardiovascular system & hematology ,Machine learning ,computer.software_genre ,Turnaround time ,Sensitivity and Specificity ,blood laboratory tests ,Machine Learning ,03 medical and health sciences ,0302 clinical medicine ,COVID-19 Testing ,blood laboratory test ,Medicine ,Humans ,030304 developmental biology ,Alternative methods ,0303 health sciences ,Training set ,Hematologic Tests ,medicine.diagnostic_test ,Receiver operating characteristic ,business.industry ,SARS-CoV-2 ,Biochemistry (medical) ,External validation ,Complete blood count ,COVID-19 ,General Medicine ,Gold standard (test) ,Triage ,Blood Cell Count ,machine learning ,Area Under Curve ,Artificial intelligence ,business ,computer ,Algorithms ,Blood Chemical Analysis ,gradient boosted decision tree - Abstract
Objectives The rRT-PCR test, the current gold standard for the detection of coronavirus disease (COVID-19), presents with known shortcomings, such as long turnaround time, potential shortage of reagents, false-negative rates around 15–20%, and expensive equipment. The hematochemical values of routine blood exams could represent a faster and less expensive alternative. Methods Three different training data set of hematochemical values from 1,624 patients (52% COVID-19 positive), admitted at San Raphael Hospital (OSR) from February to May 2020, were used for developing machine learning (ML) models: the complete OSR dataset (72 features: complete blood count (CBC), biochemical, coagulation, hemogasanalysis and CO-Oxymetry values, age, sex and specific symptoms at triage) and two sub-datasets (COVID-specific and CBC dataset, 32 and 21 features respectively). 58 cases (50% COVID-19 positive) from another hospital, and 54 negative patients collected in 2018 at OSR, were used for internal-external and external validation. Results We developed five ML models: for the complete OSR dataset, the area under the receiver operating characteristic curve (AUC) for the algorithms ranged from 0.83 to 0.90; for the COVID-specific dataset from 0.83 to 0.87; and for the CBC dataset from 0.74 to 0.86. The validations also achieved good results: respectively, AUC from 0.75 to 0.78; and specificity from 0.92 to 0.96. Conclusions ML can be applied to blood tests as both an adjunct and alternative method to rRT-PCR for the fast and cost-effective identification of COVID-19-positive patients. This is especially useful in developing countries, or in countries facing an increase in contagions.
- Published
- 2020