Artificial intelligence in pulmonary embolism: a bibliometric analysis of global research trends and future perspectives

Introduction Pulmonary embolism (PE) remains a leading cause of cardiovascular morbidity and death worldwide. A growing body of research has investigated artificial intelligence (AI) for the diagnosis, risk stratification and prognostication of PE, yet the global research landscape of this field has not been systematically mapped. Methods This study included 590 relevant articles on AI in the field of PE from the Web of Science Core Collection and Scopus databases. Bibliometric methods, combined with R software, VOSviewer, and CiteSpace, were used to perform a visual analysis of the retrieved data. Results Annual output grew slowly between 2006 and 2017 (62 articles, 10.5% of the total) and accelerated markedly from 2018 onwards; 375 articles (63.6%) were published between 2023 and 2025, and output in 2025 was approximately 14 times that in 2017 (exponential fit, R 2 = 0.876). The 590 articles involved 3,844 authors and 1,377 institutions from 63 countries/regions. The United States (257 articles, 9,791 citations) and China (117 articles, 1,320 citations) were the most productive countries, whereas citations per article were highest for Germany and the United States among the top 10. Author keywords clustered around diagnosis and imaging of PE, and recent burst keywords were “prediction” and “risk factors”. Conclusion Bibliometric mapping of 590 articles published between 2006 and 2025 shows that research on AI in PE has entered a phase of rapid growth, is dominated by a small number of countries and institutions, and remains concentrated on imaging-based diagnosis rather than on validated clinical deployment. Although several models have shown promising discriminative performance in single- or few-centre studies, the available evidence does not yet establish clinical effectiveness or improved patient outcomes. Prospective, multicentre and externally validated studies, together with multidisciplinary and cross-national collaboration, are therefore required before AI tools can be integrated into routine PE care.

Authors

Institutions

Publication Details

Journal
Frontiers in Medicine
Published
2026-11-17
DOI
https://doi.org/10.3389/fmed.2026.1978579
Primary Topic
Venous Thromboembolism Diagnosis and Management
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Artificial intelligence in pulmonary embolism: a bibliometric analysis of global research trends and future perspectives

Yulin Zhang, Kadeyanmu Abulimiti, Lin Shi, Tianping Xiong
2 links
Frontiers in Medicine
Venous Thromboembolism Diagnosis and Management
article

Artificial intelligence in pulmonary embolism: a bibliometric analysis of global research trends and future perspectives

Yulin Zhang, Kadeyanmu Abulimiti, Lin Shi, Tianping Xiong
article en

Abstract

Introduction Pulmonary embolism (PE) remains a leading cause of cardiovascular morbidity and death worldwide. A growing body of research has investigated artificial intelligence (AI) for the diagnosis, risk stratification and prognostication of PE, yet the global research landscape of this field has not been systematically mapped. Methods This study included 590 relevant articles on AI in the field of PE from the Web of Science Core Collection and Scopus databases. Bibliometric methods, combined with R software, VOSviewer, and CiteSpace, were used to perform a visual analysis of the retrieved data. Results Annual output grew slowly between 2006 and 2017 (62 articles, 10.5% of the total) and accelerated markedly from 2018 onwards; 375 articles (63.6%) were published between 2023 and 2025, and output in 2025 was approximately 14 times that in 2017 (exponential fit, R 2 = 0.876). The 590 articles involved 3,844 authors and 1,377 institutions from 63 countries/regions. The United States (257 articles, 9,791 citations) and China (117 articles, 1,320 citations) were the most productive countries, whereas citations per article were highest for Germany and the United States among the top 10. Author keywords clustered around diagnosis and imaging of PE, and recent burst keywords were “prediction” and “risk factors”. Conclusion Bibliometric mapping of 590 articles published between 2006 and 2025 shows that research on AI in PE has entered a phase of rapid growth, is dominated by a small number of countries and institutions, and remains concentrated on imaging-based diagnosis rather than on validated clinical deployment. Although several models have shown promising discriminative performance in single- or few-centre studies, the available evidence does not yet establish clinical effectiveness or improved patient outcomes. Prospective, multicentre and externally validated studies, together with multidisciplinary and cross-national collaboration, are therefore required before AI tools can be integrated into routine PE care.

Frontiers in MedicineVol. 13
First Affiliated Hospital of Xinjiang Medical University (CN), Xintai People's Hospital (CN), China Medical University (CN)
Good health and well-being
Venous Thromboembolism Diagnosis and Management
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon