EUVIMEDEUVIMED entdecken
Multizentrische StudieVergleichsstudieOpen Access · CC BY 4.0

Assessing the robustness of an artificial intelligence segmentation model for quantitative cardiovascular magnetic resonance imaging across cardiac phenotypes

Hadil Saad, Clemens Ammann, Thomas Hadler, Yashraj Bhoyroo, Philine Reisdorf, Jana Veit, Teodora Chitiboi, Jens Wetzl, Christian Geppert, Jeanette Schulz-Menger

The International Journal of Cardiovascular Imaging · 2025

DOI: 10.1007/s10554-025-03596-3 ↗

Kurz gefasst

Multizentrische Studie in der Zeitschrift The International Journal of Cardiovascular Imaging, 2025. Verfasst von Hadil Saad und 9 weiteren Autor:innen. Beteiligte Einrichtung: Charité - Universitätsmedizin Berlin, corporate member of Freie Universität Berlin and Humboldt Universität zu Berlin, Berlin, Germany. Hauptthemen laut MEDLINE: Image Interpretation, Computer-Assisted, Ventricular Function, Left, Heart Diseases, Artificial Intelligence. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Charité - Universitätsmedizin Berlin.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Purpose

To introduce an artificial intelligence-based cardiovascular magnetic resonance segmentation algorithm (Nick) for automated quantification of function and parametric mapping across cardiac phenotypes reflecting clinical routine.

Methods

Nick was compared to manual gold standard (GS) segmentations in 359 multi-centre cases at 1.5T and 3T, consisting of 104 healthy individuals and 255 diseased patients with various cardiac phenotypes. Left and right ventricular (LV, RV) volumes and LV mass (LVM) were derived from short-axis segmentations. For parametric mapping, the LV myocardium was segmented to quantify T1 and T2 relaxation times. Statistical analysis comprised mean differences, correlation coefficients (R²), Bland-Altman analysis, tolerance range assessments, and paired boxplots. The number of slices and contours requiring manual correction was estimated based on slice-level differences.

Results

Nick demonstrated high agreement with the GS for LV and RV volume estimations (R²≥0.93) and LVM quantification (R²=0.86). For the ejection fractions, correlations were slightly lower (R²=0.85/0.72 for LV/RV) with small mean differences (+ 1.14%/-2.48% for LV/RV). T1 and T2 mapping values showed excellent agreement with manual reference values (R²≥0.92) and minimal biases (-1.64/0.14 ms for T1/T2). Nick underestimated LV volumes at end-diastole (-4.48 ml) and end-systole (-3.28 ml) as well as the RV end-diastolic volume (-5.14 ml) and stroke volume (-6.75 ml). Nonetheless, tolerance testing for mean deviations revealed clinically acceptable biases for all comparisons, and less than two slices per case required correction on average.

Conclusion

Comparison to expert segmentations revealed robust performance of Nick in routine clinical cases with variable pathology, supporting its future integration into clinical workflows. Graphical Abstract

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Image Interpretation, Computer-AssistedVentricular Function, LeftHeart DiseasesArtificial IntelligenceMagnetic Resonance Imaging, CineHumansPredictive Value of TestsMalePhenotypeMiddle AgedFemaleReproducibility of ResultsAdultStroke VolumeAgedVentricular Function, RightAlgorithmsCase-Control StudiesYoung AdultAutomation

Stichwörter der Autor:innen

Artificial intelligenceCardiac functionCardiac phenotypesCardiovascular magnetic resonanceSegmentationTissue characterization

Autor:innen und Institutionen

  • Charité - Universitätsmedizin Berlin, corporate member of Freie Universität Berlin and Humboldt Universität zu Berlin, Berlin, Germany.
  • Working Group on CMR, Experimental and Clinical Research Center, Max Delbrück Center for Molecular Medicine in the Helmholtz Association and Charité - Universitätsmedizin Berlin, Berlin, Germany.
  • Department of Cardiology and Nephrology, Helios Hospital Berlin-Buch, Berlin, Germany.
  • DZHK (German Centre for Cardiovascular Research), partner site Berlin, Berlin, Germany.
  • Siemens Healthcare GmbH, Hamburg, Germany.
  • Research & Clinical Translation, Magnetic Resonance, Siemens Healthineers AG, Erlangen, Germany.

Förderung

  • Charité - Universitätsmedizin Berlin

Zitieren

APA

Saad, H., Ammann, C., Hadler, T., Bhoyroo, Y., Reisdorf, P., Veit, J., Chitiboi, T., Wetzl, J., Geppert, C., & Schulz-Menger, J. (2025). Assessing the robustness of an artificial intelligence segmentation model for quantitative cardiovascular magnetic resonance imaging across cardiac phenotypes. The International Journal of Cardiovascular Imaging. https://doi.org/10.1007/s10554-025-03596-3

Vancouver

Saad H, Ammann C, Hadler T, Bhoyroo Y, Reisdorf P, Veit J, et al. Assessing the robustness of an artificial intelligence segmentation model for quantitative cardiovascular magnetic resonance imaging across cardiac phenotypes. The International Journal of Cardiovascular Imaging. 2025. doi:10.1007/s10554-025-03596-3

BibTeX
@article{Saad2025,
  author = {Saad, Hadil and Ammann, Clemens and Hadler, Thomas and Bhoyroo, Yashraj and Reisdorf, Philine and Veit, Jana and Chitiboi, Teodora and Wetzl, Jens and Geppert, Christian and Schulz-Menger, Jeanette},
  title = {Assessing the robustness of an artificial intelligence segmentation model for quantitative cardiovascular magnetic resonance imaging across cardiac phenotypes},
  journal = {The International Journal of Cardiovascular Imaging},
  year = {2025},
  doi = {10.1007/s10554-025-03596-3}
}
Kostenlos starten →