Which explanations do clinicians prefer? A comparative evaluation of XAI understandability and actionability in predicting the need for hospitalization.

Journal: BMC medical informatics and decision making
Published Date: Jul 16, 2025

Abstract

BACKGROUND: This study aims to address the gap in understanding clinicians' attitudes toward explainable AI (XAI) methods applied to machine learning models using tabular data, commonly found in clinical settings. It specifically explores clinicians' perceptions of different XAI methods from the ALFABETO project, which predicts COVID-19 patient hospitalization based on clinical, laboratory, and chest X-ray at time of presentation to the Emergency Department. The focus is on two cognitive dimensions: understandability and actionability of the explanations provided by explainable-by-design and post-hoc methods.

Authors

  • Laura Bergomi
    Department of Electrical, Computer and Biomedical Engineering, University of Pavia, Pavia, Italy. Electronic address: laura.bergomi01@universitadipavia.it.
  • Giovanna Nicora
    Department of Electrical, Computer and Biomedical Engineering, University of Pavia, Pavia, Italy.
  • Marta Anna Orlowska
    Department of Electrical, Computer and Biomedical Engineering, Via Ferrata 5, Pavia, 27100, Italy.
  • Chiara Podrecca
    Department of Electrical, Computer and Biomedical Engineering, Via Ferrata 5, Pavia, 27100, Italy.
  • Riccardo Bellazzi
    Department of Electrical, Computer and Biomedical Engineering, University of Pavia, Pavia, Italy.
  • Caterina Fregosi
    Department of Informatics, System and Communication, University of Milan-Bicocca, Milan, Italy.
  • Francesco Salinaro
    Emergency Department, IRCCS Policlinico San Matteo Foundation, Pavia, Italy.
  • Marco Bonzano
    Emergency Medicine Unit and Emergency Medicine Postgraduate Training Program, Department of Internal Medicine, IRCCS Policlinico San Matteo Foundation, University of Pavia, Pavia, Italy.
  • Giuseppe Crescenzi
    Emergency Medicine Unit and Emergency Medicine Postgraduate Training Program, Department of Internal Medicine, IRCCS Policlinico San Matteo Foundation, University of Pavia, Pavia, Italy.
  • Francesco Speciale
    Emergency Medicine Unit and Emergency Medicine Postgraduate Training Program, Department of Internal Medicine, IRCCS Policlinico San Matteo Foundation, University of Pavia, Pavia, Italy.
  • Santi Di Pietro
    Emergency Medicine Unit and Emergency Medicine Postgraduate Training Program, Department of Internal Medicine, IRCCS Policlinico San Matteo Foundation, University of Pavia, Pavia, Italy.
  • Valentina Zuccaro
    Infectious Diseases Unit, Fondazione IRCCS Policlinico San Matteo, 27100 Pavia, Italy.
  • Erika Asperges
    Infectious Diseases Unit, Fondazione IRCCS Policlinico di Pavia, Pavia, Italy.
  • Paolo Sacchi
    Infectious Diseases Unit, Fondazione IRCCS Policlinico di Pavia, Pavia, Italy.
  • Pietro Valsecchi
    Infectious Diseases Unit, Fondazione IRCCS Policlinico di Pavia, Pavia, Italy.
  • Elisabetta Pagani
    Infectious Diseases Unit, Fondazione IRCCS Policlinico San Matteo, 27100 Pavia, Italy.
  • Michele Catalano
    Department of Clinical, Surgical, Diagnostic and Pediatric Sciences, University of Pavia, Pavia, Italy.
  • Chandra Bortolotto
    Unit of Imaging and Radiotherapy, Department of Clinical-Surgical, Diagnostic and Pediatric Sciences, University of Pavia, Pavia, Italy.
  • Lorenzo Preda
    Unit of Imaging and Radiotherapy, Department of Clinical-Surgical, Diagnostic and Pediatric Sciences, University of Pavia, Pavia, Italy.
  • Enea Parimbelli
    Telfer School of Management, University of Ottawa, Ottawa, ON, Canada.

Keywords

External Resources

Popular Topics
Recent Journals