EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Incidental pulmonary embolism detection on routine contrast-enhanced computed tomography before and after deployment of a commercial artificial intelligence tool: a real-world evaluation

Fergus O'Herlihy iD, Dora Gorman, Lara Toerien, Fionnuala Ni Ainle, Brian Gibney, Peter MacMahon

British Journal of Radiology · 2026

DOI: 10.1093/bjr/tqag222 ↗

Kurz gefasst

Fachartikel in der Zeitschrift British Journal of Radiology, 2026. Verfasst von Fergus O'Herlihy und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, Mater Misericordiae University Hospital, Dublin D07, Ireland. Hauptthemen laut MEDLINE: Pulmonary Embolism, Artificial Intelligence, Contrast Media, Tomography, X-Ray Computed. Die Arbeit nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objectives

To assess the impact of deploying AI software for pulmonary embolism (PE) detection on reporting rates of incidental PE (iPE) on routine contrast‑enhanced CT.

Methods

Consecutive adult patients undergoing contrast-enhanced CT in a large tertiary care hospital were included. Two time periods were compared: September 1-December 31, 2022 (before the introduction of dedicated artificial intelligence (AI) tools for PE detection) and September 1-December 31, 2023 (following introduction of AI). Cases positive for PE were identified retrospectively. Data on baseline demographics, imaging variables, and immediate clinical management were gathered from the electronic health record.

Results

A total of 11 690 contrast-enhanced CT studies were analyzed, 10 411 of which were eligible non-computed tomography pulmonary angiography (non-CTPA) studies. iPE detection rate increased from 0.35% to 0.63% (P = .0389) following AI implementation. The segmental iPE detection rate increased significantly (0.08% vs 0.42%, P < .001). In the post‑AI period, the iPE algorithm did not analyze 21.3% of non‑CTPA studies. AI sensitivity for iPE detection was 65.6% among analyzed studies and 60.0% when all eligible examinations were included. AI sensitivity was 71.4%, 77.8%, and 56.2% for early arterial, late arterial, and venous phase scans, respectively.

Conclusions

The introduction of a dedicated AI tool for iPE detection significantly increased iPE detection, particularly segmental iPE. This occurred despite incomplete AI analysis of eligible scans and a modest AI algorithm sensitivity. The proportion of eligible examinations processed by AI and algorithm sensitivity varied by CT protocol and contrast phase, highlighting the importance of evaluating and monitoring AI tools under real-world clinical conditions following implementation. Advances in knowledge Deployment of a commercial PE AI detection system in routine practice was associated with a significant increase in reported iPE on non‑CTPA CT, predominantly at the segmental level. However, the AI analyzed only 78.7% of eligible non-CTPA examinations, highlighting incomplete examination coverage as an important real-world implementation limitation. AI coverage and performance also varied across CT protocols and contrast phases, underscoring the importance of protocol-specific evaluation following clinical deployment.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Pulmonary EmbolismArtificial IntelligenceContrast MediaTomography, X-Ray ComputedIncidental FindingsHumansFemaleRetrospective StudiesMiddle AgedMaleAgedSensitivity and SpecificityAdultAlgorithms

Stichwörter der Autor:innen

artificial intelligenceincidencepulmonary embolus

Substanzen

Contrast Media

Autor:innen und Institutionen

  • Department of Radiology, Mater Misericordiae University Hospital, Dublin D07, Ireland.
  • Department of Haematology, Mater Misericordiae University Hospital, University College Dublin, Dublin D04, Ireland.
  • School of Medicine, University College Dublin, Dublin D04, Ireland.

ORCID-Profile: Fergus O'Herlihy

Zitieren

APA

O'Herlihy, F., Gorman, D., Toerien, L., Ni Ainle, F., Gibney, B., & MacMahon, P. (2026). Incidental pulmonary embolism detection on routine contrast-enhanced computed tomography before and after deployment of a commercial artificial intelligence tool: a real-world evaluation. British Journal of Radiology. https://doi.org/10.1093/bjr/tqag222

Vancouver

O'Herlihy F, Gorman D, Toerien L, Ni Ainle F, Gibney B, MacMahon P. Incidental pulmonary embolism detection on routine contrast-enhanced computed tomography before and after deployment of a commercial artificial intelligence tool: a real-world evaluation. British Journal of Radiology. 2026. doi:10.1093/bjr/tqag222

BibTeX
@article{OHerlihy2026,
  author = {O'Herlihy, Fergus and Gorman, Dora and Toerien, Lara and Ni Ainle, Fionnuala and Gibney, Brian and MacMahon, Peter},
  title = {Incidental pulmonary embolism detection on routine contrast-enhanced computed tomography before and after deployment of a commercial artificial intelligence tool: a real-world evaluation},
  journal = {British Journal of Radiology},
  year = {2026},
  doi = {10.1093/bjr/tqag222}
}
Kostenlos starten →