EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep learning segmentation model for quantification of infarct size in pigs with myocardial ischemia/reperfusion

Felix Braczko iD, Andreas Skyschally iD, Helmut Lieder iD, Jakob Nikolas Kather iD, Petra Kleinbongard iD, Gerd Heusch iD

Basic Research in Cardiology · 30.09.2024 · 119 (6) · 923-936

DOI: 10.1007/s00395-024-01081-x ↗

Kurz gefasst

Fachartikel in der Zeitschrift Basic Research in Cardiology, online erschienen am 30.09.2024. Verfasst von Felix Braczko und 5 weiteren Autor:innen. Beteiligte Einrichtung: Institute for Pathophysiology, West German Heart and Vascular Center, University of Duisburg-Essen, Hufelandstr. 55, 45122, Essen, Germany. Hauptthemen laut MEDLINE: Deep Learning, Myocardial Reperfusion Injury, Disease Models, Animal, Myocardial Infarction. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 68 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: European Union COST Action EU-METAHEART, European Union COST Action CA-METAHEART, European Union COST Action CARDIOPROTECTION.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Infarct size (IS) is the most robust end point for evaluating the success of preclinical studies on cardioprotection. The gold standard for IS quantification in ischemia/reperfusion (I/R) experiments is triphenyl tetrazolium chloride (TTC) staining, typically done manually. This study aimed to determine if automation through deep learning segmentation is a time-saving and valid alternative to standard IS quantification. High-resolution images from TTC-stained, macroscopic heart slices were retrospectively collected from pig experiments ( n = 390) with I/R without/with cardioprotection to cover a wide IS range. Existing IS data from pig experiments, quantified using a standard method of manual and subsequent digital labeling of film-scan annotations, were used as reference. To automate the evaluation process with the aim to be more objective and save time, a deep learning pipeline was implemented; the collected images ( n = 3869) were pre-processed by cropping and labeled (image annotations). To ensure their usability as training data for a deep learning segmentation model, IS was quantified from image annotations and compared to IS quantified using the existing film-scan annotations. A supervised deep learning segmentation model based on dynamic U-Net architecture was developed and trained. The evaluation of the trained model was performed by fivefold cross-validation ( n = 220 experiments) and testing on an independent test set ( n = 170 experiments). Performance metrics (Dice similarity coefficient [DSC], pixel accuracy [ACC], average precision [mAP]) were calculated. IS was then quantified from predictions and compared to IS quantified from image annotations (linear regression, Pearson’s r ; analysis of covariance; Bland–Altman plots). Performance metrics near 1 indicated a strong model performance on cross-validated data (DSC: 0.90, ACC: 0.98, mAP: 0.90) and on the test set data (DSC: 0.89, ACC: 0.98, mAP: 0.93). IS quantified from predictions correlated well with IS quantified from image annotations in all data sets (cross-validation: r = 0.98; test data set: r = 0.95) and analysis of covariance identified no significant differences. The model reduced the IS quantification time per experiment from approximately 90 min to 20 s. The model was further tested on a preliminary test set from experiments in isolated, saline-perfused rat hearts with regional I/R without/with cardioprotection ( n = 27). There was also no significant difference in IS between image annotations and predictions, but the performance on the test set data from rat hearts was lower (DSC: 0.66, ACC: 0.91, mAP: 0.65). IS quantification using a deep learning segmentation model is a valid and time-efficient alternative to manual and subsequent digital labeling.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningMyocardial Reperfusion InjuryDisease Models, AnimalMyocardial InfarctionAnimalsSwineMyocardiumRetrospective StudiesReproducibility of Results

Stichwörter der Autor:innen

Deep learningInfarct sizeMyocardial ischemia/reperfusionPigSegmentation

Autor:innen und Institutionen

  • Institute for Pathophysiology, West German Heart and Vascular Center, University of Duisburg-Essen, Hufelandstr. 55, 45122, Essen, Germany.
  • Department of Medicine III, University Hospital RWTH Aachen, Aachen, Germany.
  • Else Kroener Fresenius Center for Digital Health, Medical Faculty Carl Gustav Carus, Technical University Dresden, Dresden, Germany.
  • Institute for Pathophysiology, West German Heart and Vascular Center, University of Duisburg-Essen, Hufelandstr. 55, 45122, Essen, Germany. gerd.heusch@uk-essen.de.

ORCID-Profile: Felix Braczko, Andreas Skyschally, Helmut Lieder, Jakob Nikolas Kather, Petra Kleinbongard, Gerd Heusch

Förderung

  • European Union COST Action EU-METAHEART
  • European Union COST Action CA-METAHEART
  • European Union COST Action CARDIOPROTECTION
  • Deutsche Forschungsgemeinschaft
  • GSK
  • Universitätsklinikum Essen

Zitieren

APA

Braczko, F., Skyschally, A., Lieder, H., Kather, J. N., Kleinbongard, P., & Heusch, G. (2024). Deep learning segmentation model for quantification of infarct size in pigs with myocardial ischemia/reperfusion. Basic Research in Cardiology, 119(6), 923-936. https://doi.org/10.1007/s00395-024-01081-x

Vancouver

Braczko F, Skyschally A, Lieder H, Kather JN, Kleinbongard P, Heusch G. Deep learning segmentation model for quantification of infarct size in pigs with myocardial ischemia/reperfusion. Basic Research in Cardiology. 2024;119(6):923-936. doi:10.1007/s00395-024-01081-x

BibTeX
@article{Braczko2024,
  author = {Braczko, Felix and Skyschally, Andreas and Lieder, Helmut and Kather, Jakob Nikolas and Kleinbongard, Petra and Heusch, Gerd},
  title = {Deep learning segmentation model for quantification of infarct size in pigs with myocardial ischemia/reperfusion},
  journal = {Basic Research in Cardiology},
  year = {2024},
  volume = {119},
  number = {6},
  pages = {923--936},
  doi = {10.1007/s00395-024-01081-x}
}
Kostenlos starten →