Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Performance of a point-of-care ultrasound platform for artificial intelligence-enabled assessment of pulmonary B-lines

Labaf, Ashkan LU ; Åhman-Persson, Linda ; Husu, Leo Silvén ; Smith, J. Gustav LU orcid ; Ingvarsson, Annika LU orcid and Evaldsson, Anna Werther LU orcid (2025) In Cardiovascular Ultrasound 23(1).
Abstract

Background: The incorporation of artificial intelligence (AI) into point-of-care ultrasound (POCUS) platforms has rapidly increased. The number of B-lines present on lung ultrasound (LUS) serve as a useful tool for the assessment of pulmonary congestion. Interpretation, however, requires experience and therefore AI automation has been pursued. This study aimed to test the agreement between the AI software embedded in a major vendor POCUS system and visual expert assessment. Methods: This single-center prospective study included 55 patients hospitalized for various respiratory symptoms, predominantly acutely decompensated heart failure. A 12-zone protocol was used. Two experts in LUS independently categorized B-lines into 0, 1–2, 3–4,... (More)

Background: The incorporation of artificial intelligence (AI) into point-of-care ultrasound (POCUS) platforms has rapidly increased. The number of B-lines present on lung ultrasound (LUS) serve as a useful tool for the assessment of pulmonary congestion. Interpretation, however, requires experience and therefore AI automation has been pursued. This study aimed to test the agreement between the AI software embedded in a major vendor POCUS system and visual expert assessment. Methods: This single-center prospective study included 55 patients hospitalized for various respiratory symptoms, predominantly acutely decompensated heart failure. A 12-zone protocol was used. Two experts in LUS independently categorized B-lines into 0, 1–2, 3–4, and ≥ 5. The intraclass correlation coefficient (ICC) was used to determine agreement. Results: A total of 672 LUS zones were obtained, with 584 (87%) eligible for analysis. Compared with expert reviewers, the AI significantly overcounted number of B-lines per patient (23.5 vs. 2.8, p < 0.001). A greater proportion of zones with > 5 B-lines was found by the AI than by the reviewers (38% vs. 4%, p < 0.001). The ICC between the AI and reviewers was 0.28 for the total sum of B-lines and 0.37 for the zone-by-zone method. The interreviewer agreement was excellent, with ICCs of 0.92 and 0.91, respectively. Conclusion: This study demonstrated excellent interrater reliability of B-line counts from experts but poor agreement with the AI software embedded in a major vendor system, primarily due to overcounting. Our findings indicate that further development is needed to increase the accuracy of AI tools in LUS.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Artificial intelligence, B-lines, Lung ultrasound, POCUS
in
Cardiovascular Ultrasound
volume
23
issue
1
article number
3
publisher
BioMed Central (BMC)
external identifiers
  • pmid:40025516
  • scopus:85219638236
ISSN
1476-7120
DOI
10.1186/s12947-025-00338-2
language
English
LU publication?
yes
additional info
Publisher Copyright: © The Author(s) 2025.
id
416e0fbc-2063-4584-9d02-3faa27074721
date added to LUP
2025-03-17 12:39:29
date last changed
2025-08-05 01:06:31
@article{416e0fbc-2063-4584-9d02-3faa27074721,
  abstract     = {{<p>Background: The incorporation of artificial intelligence (AI) into point-of-care ultrasound (POCUS) platforms has rapidly increased. The number of B-lines present on lung ultrasound (LUS) serve as a useful tool for the assessment of pulmonary congestion. Interpretation, however, requires experience and therefore AI automation has been pursued. This study aimed to test the agreement between the AI software embedded in a major vendor POCUS system and visual expert assessment. Methods: This single-center prospective study included 55 patients hospitalized for various respiratory symptoms, predominantly acutely decompensated heart failure. A 12-zone protocol was used. Two experts in LUS independently categorized B-lines into 0, 1–2, 3–4, and ≥ 5. The intraclass correlation coefficient (ICC) was used to determine agreement. Results: A total of 672 LUS zones were obtained, with 584 (87%) eligible for analysis. Compared with expert reviewers, the AI significantly overcounted number of B-lines per patient (23.5 vs. 2.8, p &lt; 0.001). A greater proportion of zones with &gt; 5 B-lines was found by the AI than by the reviewers (38% vs. 4%, p &lt; 0.001). The ICC between the AI and reviewers was 0.28 for the total sum of B-lines and 0.37 for the zone-by-zone method. The interreviewer agreement was excellent, with ICCs of 0.92 and 0.91, respectively. Conclusion: This study demonstrated excellent interrater reliability of B-line counts from experts but poor agreement with the AI software embedded in a major vendor system, primarily due to overcounting. Our findings indicate that further development is needed to increase the accuracy of AI tools in LUS.</p>}},
  author       = {{Labaf, Ashkan and Åhman-Persson, Linda and Husu, Leo Silvén and Smith, J. Gustav and Ingvarsson, Annika and Evaldsson, Anna Werther}},
  issn         = {{1476-7120}},
  keywords     = {{Artificial intelligence; B-lines; Lung ultrasound; POCUS}},
  language     = {{eng}},
  number       = {{1}},
  publisher    = {{BioMed Central (BMC)}},
  series       = {{Cardiovascular Ultrasound}},
  title        = {{Performance of a point-of-care ultrasound platform for artificial intelligence-enabled assessment of pulmonary B-lines}},
  url          = {{http://dx.doi.org/10.1186/s12947-025-00338-2}},
  doi          = {{10.1186/s12947-025-00338-2}},
  volume       = {{23}},
  year         = {{2025}},
}