In this article we discuss the development of prognostic Machine Learning (ML) models for COVID-19 progression: specifically, we address the task of predicting intensive care unit (ICU) admission in the next 5 days. We developed three ML models on the basis of 4995 Complete Blood Count (CBC) tests. We propose three ML models that differ in terms of interpretability: two fully interpretable models and a black-box one. We report an AUC of. 81 and. 83 for the interpretable models (the decision tree and logistic regression, respectively), and an AUC of. 88 for the black-box model (an ensemble). This shows that CBC data and ML methods can be used for cost-effective prediction of ICU admission of COVID-19 patients: in particular, as the CBC can be acquired rapidly through routine blood exams, our models could also be applied in resource-limited settings and to get fast indications at triage and daily rounds.

Famiglini, L., Bini, G., Carobene, A., Campagner, A., Cabitza, F. (2021). Prediction of ICU admission for COVID-19 patients: A machine learning approach based on complete blood count data. In 34th IEEE International Symposium on Computer-Based Medical Systems, CBMS 2021 (pp.160-165). Institute of Electrical and Electronics Engineers Inc. [10.1109/CBMS52027.2021.00065].

Prediction of ICU admission for COVID-19 patients: A machine learning approach based on complete blood count data

Famiglini L.;Campagner A.;Cabitza F.
2021

Abstract

In this article we discuss the development of prognostic Machine Learning (ML) models for COVID-19 progression: specifically, we address the task of predicting intensive care unit (ICU) admission in the next 5 days. We developed three ML models on the basis of 4995 Complete Blood Count (CBC) tests. We propose three ML models that differ in terms of interpretability: two fully interpretable models and a black-box one. We report an AUC of. 81 and. 83 for the interpretable models (the decision tree and logistic regression, respectively), and an AUC of. 88 for the black-box model (an ensemble). This shows that CBC data and ML methods can be used for cost-effective prediction of ICU admission of COVID-19 patients: in particular, as the CBC can be acquired rapidly through routine blood exams, our models could also be applied in resource-limited settings and to get fast indications at triage and daily rounds.
paper
Complete Blood Count; COVID-19; eXplainable AI; Machine Learning; Prognosis
English
34th IEEE International Symposium on Computer-Based Medical Systems, CBMS 2021 - JUN 07-09, 2021
2021
34th IEEE International Symposium on Computer-Based Medical Systems, CBMS 2021
9781665441216
2021
2021-
160
165
9474781
open
Famiglini, L., Bini, G., Carobene, A., Campagner, A., Cabitza, F. (2021). Prediction of ICU admission for COVID-19 patients: A machine learning approach based on complete blood count data. In 34th IEEE International Symposium on Computer-Based Medical Systems, CBMS 2021 (pp.160-165). Institute of Electrical and Electronics Engineers Inc. [10.1109/CBMS52027.2021.00065].
File in questo prodotto:
File Dimensione Formato  
CBMS21_prognostico(1).pdf

accesso aperto

Tipologia di allegato: Submitted Version (Pre-print)
Dimensione 1.3 MB
Formato Adobe PDF
1.3 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/324835
Citazioni
  • Scopus 18
  • ???jsp.display-item.citation.isi??? 12
Social impact