Cargando…

Artificial Intelligence-Assisted Chest X-ray for the Diagnosis of COVID-19: A Systematic Review and Meta-Analysis

Because it is an accessible and routine image test, medical personnel commonly use a chest X-ray for COVID-19 infections. Artificial intelligence (AI) is now widely applied to improve the precision of routine image tests. Hence, we investigated the clinical merit of the chest X-ray to detect COVID-1...

Descripción completa

Detalles Bibliográficos
Autores principales: Tzeng, I-Shiang, Hsieh, Po-Chun, Su, Wen-Lin, Hsieh, Tsung-Han, Chang, Sheng-Chang
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9955250/
https://www.ncbi.nlm.nih.gov/pubmed/36832072
http://dx.doi.org/10.3390/diagnostics13040584
_version_ 1784894306851487744
author Tzeng, I-Shiang
Hsieh, Po-Chun
Su, Wen-Lin
Hsieh, Tsung-Han
Chang, Sheng-Chang
author_facet Tzeng, I-Shiang
Hsieh, Po-Chun
Su, Wen-Lin
Hsieh, Tsung-Han
Chang, Sheng-Chang
author_sort Tzeng, I-Shiang
collection PubMed
description Because it is an accessible and routine image test, medical personnel commonly use a chest X-ray for COVID-19 infections. Artificial intelligence (AI) is now widely applied to improve the precision of routine image tests. Hence, we investigated the clinical merit of the chest X-ray to detect COVID-19 when assisted by AI. We used PubMed, Cochrane Library, MedRxiv, ArXiv, and Embase to search for relevant research published between 1 January 2020 and 30 May 2022. We collected essays that dissected AI-based measures used for patients diagnosed with COVID-19 and excluded research lacking measurements using relevant parameters (i.e., sensitivity, specificity, and area under curve). Two independent researchers summarized the information, and discords were eliminated by consensus. A random effects model was used to calculate the pooled sensitivities and specificities. The sensitivity of the included research studies was enhanced by eliminating research with possible heterogeneity. A summary receiver operating characteristic curve (SROC) was generated to investigate the diagnostic value for detecting COVID-19 patients. Nine studies were recruited in this analysis, including 39,603 subjects. The pooled sensitivity and specificity were estimated as 0.9472 (p = 0.0338, 95% CI 0.9009–0.9959) and 0.9610 (p < 0.0001, 95% CI 0.9428–0.9795), respectively. The area under the SROC was 0.98 (95% CI 0.94–1.00). The heterogeneity of diagnostic odds ratio was presented in the recruited studies (I(2) = 36.212, p = 0.129). The AI-assisted chest X-ray scan for COVID-19 detection offered excellent diagnostic potential and broader application.
format Online
Article
Text
id pubmed-9955250
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-99552502023-02-25 Artificial Intelligence-Assisted Chest X-ray for the Diagnosis of COVID-19: A Systematic Review and Meta-Analysis Tzeng, I-Shiang Hsieh, Po-Chun Su, Wen-Lin Hsieh, Tsung-Han Chang, Sheng-Chang Diagnostics (Basel) Review Because it is an accessible and routine image test, medical personnel commonly use a chest X-ray for COVID-19 infections. Artificial intelligence (AI) is now widely applied to improve the precision of routine image tests. Hence, we investigated the clinical merit of the chest X-ray to detect COVID-19 when assisted by AI. We used PubMed, Cochrane Library, MedRxiv, ArXiv, and Embase to search for relevant research published between 1 January 2020 and 30 May 2022. We collected essays that dissected AI-based measures used for patients diagnosed with COVID-19 and excluded research lacking measurements using relevant parameters (i.e., sensitivity, specificity, and area under curve). Two independent researchers summarized the information, and discords were eliminated by consensus. A random effects model was used to calculate the pooled sensitivities and specificities. The sensitivity of the included research studies was enhanced by eliminating research with possible heterogeneity. A summary receiver operating characteristic curve (SROC) was generated to investigate the diagnostic value for detecting COVID-19 patients. Nine studies were recruited in this analysis, including 39,603 subjects. The pooled sensitivity and specificity were estimated as 0.9472 (p = 0.0338, 95% CI 0.9009–0.9959) and 0.9610 (p < 0.0001, 95% CI 0.9428–0.9795), respectively. The area under the SROC was 0.98 (95% CI 0.94–1.00). The heterogeneity of diagnostic odds ratio was presented in the recruited studies (I(2) = 36.212, p = 0.129). The AI-assisted chest X-ray scan for COVID-19 detection offered excellent diagnostic potential and broader application. MDPI 2023-02-05 /pmc/articles/PMC9955250/ /pubmed/36832072 http://dx.doi.org/10.3390/diagnostics13040584 Text en © 2023 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Review
Tzeng, I-Shiang
Hsieh, Po-Chun
Su, Wen-Lin
Hsieh, Tsung-Han
Chang, Sheng-Chang
Artificial Intelligence-Assisted Chest X-ray for the Diagnosis of COVID-19: A Systematic Review and Meta-Analysis
title Artificial Intelligence-Assisted Chest X-ray for the Diagnosis of COVID-19: A Systematic Review and Meta-Analysis
title_full Artificial Intelligence-Assisted Chest X-ray for the Diagnosis of COVID-19: A Systematic Review and Meta-Analysis
title_fullStr Artificial Intelligence-Assisted Chest X-ray for the Diagnosis of COVID-19: A Systematic Review and Meta-Analysis
title_full_unstemmed Artificial Intelligence-Assisted Chest X-ray for the Diagnosis of COVID-19: A Systematic Review and Meta-Analysis
title_short Artificial Intelligence-Assisted Chest X-ray for the Diagnosis of COVID-19: A Systematic Review and Meta-Analysis
title_sort artificial intelligence-assisted chest x-ray for the diagnosis of covid-19: a systematic review and meta-analysis
topic Review
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9955250/
https://www.ncbi.nlm.nih.gov/pubmed/36832072
http://dx.doi.org/10.3390/diagnostics13040584
work_keys_str_mv AT tzengishiang artificialintelligenceassistedchestxrayforthediagnosisofcovid19asystematicreviewandmetaanalysis
AT hsiehpochun artificialintelligenceassistedchestxrayforthediagnosisofcovid19asystematicreviewandmetaanalysis
AT suwenlin artificialintelligenceassistedchestxrayforthediagnosisofcovid19asystematicreviewandmetaanalysis
AT hsiehtsunghan artificialintelligenceassistedchestxrayforthediagnosisofcovid19asystematicreviewandmetaanalysis
AT changshengchang artificialintelligenceassistedchestxrayforthediagnosisofcovid19asystematicreviewandmetaanalysis