Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Jun 3;11(6):501.
doi: 10.3390/jpm11060501.

Machine Learning to Predict In-Hospital Mortality in COVID-19 Patients Using Computed Tomography-Derived Pulmonary and Vascular Features

Simone Schiaffino  1 , Marina Codari  2 , Andrea Cozzi  3 , Domenico Albano  4   5 , Marco Alì  6 , Roberto Arioli  7 , Emanuele Avola  8 , Claudio Bnà  9 , Maurizio Cariati  10 , Serena Carriero  8 , Massimo Cressoni  1 , Pietro S C Danna  7 , Gianmarco Della Pepa  8 , Giovanni Di Leo  1 , Francesco Dolci  11 , Zeno Falaschi  7 , Nicola Flor  12 , Riccardo A Foà  10   13 , Salvatore Gitto  3 , Giovanni Leati  13 , Veronica Magni  3 , Alexis E Malavazos  14 , Giovanni Mauri  15   16 , Carmelo Messina  4 , Lorenzo Monfardini  9 , Alessio Paschè  7 , Filippo Pesapane  17 , Luca M Sconfienza  3   4 , Francesco Secchi  1   3 , Edoardo Segalini  18 , Angelo Spinazzola  13 , Valeria Tombini  19 , Silvia Tresoldi  10 , Angelo Vanzulli  15   19 , Ilaria Vicentin  19 , Domenico Zagaria  7 , Dominik Fleischmann  2   20 , Francesco Sardanelli  1   3
Affiliations

Machine Learning to Predict In-Hospital Mortality in COVID-19 Patients Using Computed Tomography-Derived Pulmonary and Vascular Features

Simone Schiaffino et al. J Pers Med. .

Abstract

Pulmonary parenchymal and vascular damage are frequently reported in COVID-19 patients and can be assessed with unenhanced chest computed tomography (CT), widely used as a triaging exam. Integrating clinical data, chest CT features, and CT-derived vascular metrics, we aimed to build a predictive model of in-hospital mortality using univariate analysis (Mann-Whitney U test) and machine learning models (support vectors machines (SVM) and multilayer perceptrons (MLP)). Patients with RT-PCR-confirmed SARS-CoV-2 infection and unenhanced chest CT performed on emergency department admission were included after retrieving their outcome (discharge or death), with an 85/15% training/test dataset split. Out of 897 patients, the 229 (26%) patients who died during hospitalization had higher median pulmonary artery diameter (29.0 mm) than patients who survived (27.0 mm, p < 0.001) and higher median ascending aortic diameter (36.6 mm versus 34.0 mm, p < 0.001). SVM and MLP best models considered the same ten input features, yielding a 0.747 (precision 0.522, recall 0.800) and 0.844 (precision 0.680, recall 0.567) area under the curve, respectively. In this model integrating clinical and radiological data, pulmonary artery diameter was the third most important predictor after age and parenchymal involvement extent, contributing to reliable in-hospital mortality prediction, highlighting the value of vascular metrics in improving patient stratification.

Keywords: COVID-19; X-ray computed; computer; lung; machine learning; neural networks; prognosis; pulmonary artery; support vector machine; tomography.

PubMed Disclaimer

Conflict of interest statement

S. Schiaffino declares to have received travel support from Bracco Imaging and to be member of speakers’ bureau for General Electric Healthcare. D. Fleischmann declares to have received a research grant from Siemens AG, to be a shareholder of Segmed, Inc., and to have ownership interests in Ischemia View, Inc. F. Sardanelli declares to have received grants from or to be a member of the speakers’ bureau/advisory board for Bayer Healthcare, Bracco Group, and General Electric Healthcare. All other authors declare that they have no conflicts of interest and that they have nothing to disclose.

Figures

Figure 1
Figure 1
Example of axial slice selected at the level of pulmonary arterial main trunk bifurcation. The solid and dashed lines identify the maximum diameters of the pulmonary artery and ascending aorta, respectively.
Figure 2
Figure 2
Feature correlation matrix, the color gradient representing the Pearson’s correlation coefficient value. GGOs: ground glass opacities; PA: pulmonary artery; AA: ascending aorta.
Figure 3
Figure 3
Coefficient of the least absolute shrinkage and selection operator (LASSO) regression in training/validation data. PA: pulmonary artery; AA: ascending aorta; GGOs: ground glass opacities.
Figure 4
Figure 4
Receiver operator characteristic curves of the best support vector machine (a) and multilayer perceptron (b) models.
See this image and copyright information in PMC

References

    1. Akl E.A., Blažić I., Yaacoub S., Frija G., Chou R., Appiah J.A., Fatehi M., Flor N., Hitti E., Jafri H., et al. Use of Chest Imaging in the Diagnosis and Management of COVID-19: A WHO Rapid Advice Guide. Radiology. 2021;298:E63–E69. doi: 10.1148/radiol.2020203173. - DOI - PMC - PubMed
    1. Rubin G.D., Ryerson C.J., Haramati L.B., Sverzellati N., Kanne J.P., Raoof S., Schluger N.W., Volpi A., Yim J.-J., Martin I.B.K., et al. The Role of Chest Imaging in Patient Management during the COVID-19 Pandemic: A Multinational Consensus Statement from the Fleischner Society. Radiology. 2020;296:172–180. doi: 10.1148/radiol.2020201365. - DOI - PMC - PubMed
    1. Sverzellati N., Milanese G., Milone F., Balbi M., Ledda R.E., Silva M. Integrated Radiologic Algorithm for COVID-19 Pandemic. J. Thorac. Imaging. 2020;35:228–233. doi: 10.1097/RTI.0000000000000516. - DOI - PMC - PubMed
    1. Tang N., Li D., Wang X., Sun Z. Abnormal coagulation parameters are associated with poor prognosis in patients with novel coronavirus pneumonia. J. Thromb. Haemost. 2020;18:844–847. doi: 10.1111/jth.14768. - DOI - PMC - PubMed
    1. Nadkarni G.N., Lala A., Bagiella E., Chang H.L., Moreno P.R., Pujadas E., Arvind V., Bose S., Charney A.W., Chen M.D., et al. Anticoagulation, Bleeding, Mortality, and Pathology in Hospitalized Patients with COVID-19. J. Am. Coll. Cardiol. 2020;76:1815–1826. doi: 10.1016/j.jacc.2020.08.041. - DOI - PMC - PubMed

LinkOut - more resources