Exploring the Synergistic Potential of Radiomics and Laboratory Biomarkers for Enhanced Identification of Vulnerable COVID-19 Patients.
COVID-19
SARS-CoV-2
algorithms
cell-free nucleic acid
coronavirus infection
integrative medicine
intensive care units
thoracic radiography
Journal
Microorganisms
ISSN: 2076-2607
Titre abrégé: Microorganisms
Pays: Switzerland
ID NLM: 101625893
Informations de publication
Date de publication:
03 Jul 2023
03 Jul 2023
Historique:
received:
17
06
2023
revised:
23
06
2023
accepted:
29
06
2023
medline:
29
7
2023
pubmed:
29
7
2023
entrez:
29
7
2023
Statut:
epublish
Résumé
Severe courses and high hospitalization rates were ubiquitous during the first pandemic SARS-CoV-2 waves. Thus, we aimed to examine whether integrative diagnostics may aid in identifying vulnerable patients using crucial data and materials obtained from COVID-19 patients hospitalized between 2020 and 2021 ( Separate forward and backward feature selection was conducted for linear regression with the Intensive-Care-Unit (ICU) period as the initial target. Three-fold cross-validation was performed, and collinear parameters were reduced. The model was adapted to a logistic regression approach and verified in a validation naïve subset to avoid overfitting. The adapted integrated model classifying patients into "ICU/no ICU demand" comprises six radiomics and seven laboratory biomarkers. The models' accuracy was 0.54 for radiomics, 0.47 for cfDNA, 0.74 for routine laboratory, and 0.87 for the combined model with an AUC of 0.91. The combined model performed superior to the individual models. Thus, integrating radiomics and laboratory data shows synergistic potential to aid clinic decision-making in COVID-19 patients. Under the need for evaluation in larger cohorts, including patients with other SARS-CoV-2 variants, the identified parameters might contribute to the triage of COVID-19 patients.
Sections du résumé
BACKGROUND
BACKGROUND
Severe courses and high hospitalization rates were ubiquitous during the first pandemic SARS-CoV-2 waves. Thus, we aimed to examine whether integrative diagnostics may aid in identifying vulnerable patients using crucial data and materials obtained from COVID-19 patients hospitalized between 2020 and 2021 (
METHODS
METHODS
Separate forward and backward feature selection was conducted for linear regression with the Intensive-Care-Unit (ICU) period as the initial target. Three-fold cross-validation was performed, and collinear parameters were reduced. The model was adapted to a logistic regression approach and verified in a validation naïve subset to avoid overfitting.
RESULTS
RESULTS
The adapted integrated model classifying patients into "ICU/no ICU demand" comprises six radiomics and seven laboratory biomarkers. The models' accuracy was 0.54 for radiomics, 0.47 for cfDNA, 0.74 for routine laboratory, and 0.87 for the combined model with an AUC of 0.91.
CONCLUSION
CONCLUSIONS
The combined model performed superior to the individual models. Thus, integrating radiomics and laboratory data shows synergistic potential to aid clinic decision-making in COVID-19 patients. Under the need for evaluation in larger cohorts, including patients with other SARS-CoV-2 variants, the identified parameters might contribute to the triage of COVID-19 patients.
Identifiants
pubmed: 37512912
pii: microorganisms11071740
doi: 10.3390/microorganisms11071740
pmc: PMC10384842
pii:
doi:
Types de publication
Journal Article
Langues
eng
Subventions
Organisme : Ministry of Science, Research and Arts, Baden-Württemberg, Germany: "Measures to combat the coronavirus SARS CoV-2 pandemic in the field of medical research"
ID : special funding programme COVID-19 (Chap. 1499 TG 93)), (MA10).
Références
Radiology. 2020 Jul;296(1):172-180
pubmed: 32255413
PLoS One. 2022 Jul 29;17(7):e0271787
pubmed: 35905122
Innate Immun. 2021 Apr;27(3):240-250
pubmed: 33646058
Nat Commun. 2021 Jan 27;12(1):634
pubmed: 33504775
Front Mol Biosci. 2021 Mar 18;8:600881
pubmed: 33816549
Cytokine. 2022 Jan;149:155755
pubmed: 34773859
BMC Med Genomics. 2020 Nov 23;13(1):178
pubmed: 33228632
Crit Rev Clin Lab Sci. 2020 Jan;57(1):1-21
pubmed: 31603708
Thromb Res. 2020 Dec;196:99-105
pubmed: 32853982
Int J Med Inform. 2021 Oct;154:104545
pubmed: 34464848
Sci Rep. 2022 Mar 14;12(1):4329
pubmed: 35288579
J Transl Med. 2021 Jan 7;19(1):29
pubmed: 33413480
PLoS One. 2021 Apr 1;16(4):e0249285
pubmed: 33793600
Tumour Biol. 2019 Aug;41(8):1010428319866369
pubmed: 31402761
Abdom Radiol (NY). 2021 Apr;46(4):1651-1658
pubmed: 33098478
Insights Imaging. 2020 Aug 12;11(1):91
pubmed: 32785796
Biomed Pharmacother. 2021 Nov;143:112107
pubmed: 34488083
J Thromb Haemost. 2020 Jun;18(6):1324-1329
pubmed: 32306492
Radiol Cardiothorac Imaging. 2020 Mar 25;2(2):e200152
pubmed: 33778571
Clin Chem Lab Med. 2006;44(3):311-6
pubmed: 16519604
BMJ Open. 2021 Oct 1;11(10):e043790
pubmed: 34598979
Brief Bioinform. 2019 Mar 22;20(2):492-503
pubmed: 29045534
Sci Adv. 2016 Mar 25;2(3):e1501332
pubmed: 27051864
Int J Infect Dis. 2021 Jun;107:221-227
pubmed: 33932604
Korean J Radiol. 2021 Jun;22(6):994-1004
pubmed: 33686818
Curr Pathobiol Rep. 2021;9(4):107-117
pubmed: 34900401
Biomedicines. 2023 Apr 26;11(5):
pubmed: 37238955
Ann Med. 2021 Dec;53(1):257-266
pubmed: 33410720
Ann Lab Med. 2021 Nov 1;41(6):540-548
pubmed: 34108281
Crit Care. 2021 Apr 14;25(1):145
pubmed: 33853641
Radiology. 2004 Mar;230(3):836-44
pubmed: 14990845
Nat Med. 2020 Jul;26(7):1037-1040
pubmed: 32393804
Obesity (Silver Spring). 2020 Jul;28(7):1195-1199
pubmed: 32271993