Clinical usability of deep learning-based saliency maps for occlusion myocardial infarction identification from the prehospital 12-Lead electrocardiogram.

Journal: Journal of electrocardiology
PMID: 39255653

Abstract

INTRODUCTION: Deep learning (DL) models offer improved performance in electrocardiogram (ECG)-based classification over rule-based methods. However, for widespread adoption by clinicians, explainability methods, like saliency maps, are essential.

Authors

  • Nathan T Riek
    Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, PA, USA.
  • Tanmay A Gokhale
    Division of Cardiology, University of Pittsburgh, Pittsburgh, PA, USA; University of Pittsburgh Medical Center (UPMC), Pittsburgh, PA, USA.
  • Christian Martin-Gill
    Department of Emergency Medicine, University of Pittsburgh, Pittsburgh, PA, USA.
  • Karina Kraevsky-Philips
    Department of Acute and Tertiary Care Nursing, University of Pittsburgh, Pittsburgh, PA, USA; University of Pittsburgh Medical Center (UPMC), Pittsburgh, PA, USA.
  • Jessica K Zègre-Hemsey
    School of Nursing, University of North Carolina at Chapel Hill.
  • Samir Saba
    Division of Cardiology, School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA.
  • Clifton W Callaway
    Department of Emergency Medicine, University of Pittsburgh, Pittsburgh, PA, USA; University of Pittsburgh Medical Center (UPMC), Pittsburgh, PA, USA.
  • Murat Akcakaya
    Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, PA, USA, akcakaya@pitt.edu,tugcebusraiu@gmail.com.
  • Salah S Al-Zaiti
    Departments of Acute & Tertiary Care Nursing, Emergency Medicine, and Cardiology, University of Pittsburgh, Pittsburgh, PA, USA.

Keywords

External Resources

Popular Topics
Recent Journals