EUVIMEDΑνακαλύψτε το EUVIMED
MetaanalyseSystematische ÜbersichtOpen access

Diagnostic Test Accuracy of Deep Learning Prediction Models on COVID-19 Severity: Systematic Review and Meta-Analysis

Changyu Wang iD, Siru Liu iD, Yu Tang iD, Hao Yang iD, Jialin Liu iD

Journal of Medical Internet Research · 2023-07-21 · 25 · e46340

DOI: 10.2196/46340 ↗

At a glance

Metaanalyse in Journal of Medical Internet Research, published online on 2023-07-21. Written by Changyu Wang and 4 co-authors. Institution: Department of Medical Informatics, West China Medical School, Sichuan University, Chengdu, China. Major MEDLINE topics: COVID-19, Deep Learning. Cited 5 times so far (Crossref); 67 references. The full text is freely available in PubMed Central.

ΠερίληψηSubjectsΒιβλιογραφικά στοιχείαCiteRelated studies

Abstract (original language)

Background

Deep learning (DL) prediction models hold great promise in the triage of COVID-19.

Objective

We aimed to evaluate the diagnostic test accuracy of DL prediction models for assessing and predicting the severity of COVID-19.

Methods

We searched PubMed, Scopus, LitCovid, Embase, Ovid, and the Cochrane Library for studies published from December 1, 2019, to April 30, 2022. Studies that used DL prediction models to assess or predict COVID-19 severity were included, while those without diagnostic test accuracy analysis or severity dichotomies were excluded. QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies 2), PROBAST (Prediction Model Risk of Bias Assessment Tool), and funnel plots were used to estimate the bias and applicability.

Results

A total of 12 retrospective studies involving 2006 patients reported the cross-sectionally assessed value of DL on COVID-19 severity. The pooled sensitivity and area under the curve were 0.92 (95% CI 0.89-0.94; I2=0.00%) and 0.95 (95% CI 0.92-0.96), respectively. A total of 13 retrospective studies involving 3951 patients reported the longitudinal predictive value of DL for disease severity. The pooled sensitivity and area under the curve were 0.76 (95% CI 0.74-0.79; I2=0.00%) and 0.80 (95% CI 0.76-0.83), respectively.

Conclusions

DL prediction models can help clinicians identify potentially severe cases for early triage. However, high-quality research is lacking. Trial Registration PROSPERO CRD42022329252; https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD 42022329252

Subjects

MeSH terms (MEDLINE)

COVID-19Deep LearningHumansRetrospective StudiesPubMedDiagnostic Tests, RoutineCOVID-19 Testing

Author keywords

AICOVID-19Severity of Illness Indexaccuracydeep learningdigital health interventiondisease severitymeta-analysisprediction modelprognosisprognostics and health managementsystematic review

Authors and institutions

  • Department of Medical Informatics, West China Medical School, Sichuan University, Chengdu, China.
  • West China College of Stomatology, Sichuan University, Chengdu, China.
  • Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN, United States.
  • Xiangya School of Medicine, Central South University, Changsha, China.
  • Information Center, West China Hospital, Sichuan University, Chengdu, China.

ORCID profiles: Changyu Wang, Siru Liu, Yu Tang, Hao Yang, Jialin Liu

Cite

APA

Wang, C., Liu, S., Tang, Y., Yang, H., & Liu, J. (2023). Diagnostic Test Accuracy of Deep Learning Prediction Models on COVID-19 Severity: Systematic Review and Meta-Analysis. Journal of Medical Internet Research, 25, e46340. https://doi.org/10.2196/46340

Vancouver

Wang C, Liu S, Tang Y, Yang H, Liu J. Diagnostic Test Accuracy of Deep Learning Prediction Models on COVID-19 Severity: Systematic Review and Meta-Analysis. Journal of Medical Internet Research. 2023;25:e46340. doi:10.2196/46340

BibTeX
@article{Wang2023,
  author = {Wang, Changyu and Liu, Siru and Tang, Yu and Yang, Hao and Liu, Jialin},
  title = {Diagnostic Test Accuracy of Deep Learning Prediction Models on COVID-19 Severity: Systematic Review and Meta-Analysis},
  journal = {Journal of Medical Internet Research},
  year = {2023},
  volume = {25},
  pages = {e46340},
  doi = {10.2196/46340}
}
Start for free →