Explainable artificial intelligence in medical imaging: how to interpret, evaluate, and use artificial intelligence explanations
Society of Abdominal Radiology (SAR)
Abstract
Most artificial intelligence (AI) models used in radiology are black boxes-they produce predictions without explaining the basis of their outputs, raising concerns about clinical safety, accountability, and trust. To address this, a growing body of methods has been developed to help clinicians understand and evaluate AI predictions. This field, known as explainable AI (XAI), aims to help clinicians interrogate, interpret, and critically evaluate AI predictions by identifying factors associated with model outputs. In this educational and practical review, we provide an accessible overview of XAI tailored for practicing radiologists and physicians. We cover the major categories of explanation methods, including saliency maps, perturbation-based and feature-attribution approaches, concept- based methods, and example-based reasoning, as well as uncertainty quantification as a complementary approach for assessing prediction reliability, along with common misconceptions and emerging regulatory obligations. We aim to make XAI easier for healthcare professionals to understand, as effective oversight of AI tools has become a core competency for the modern radiologist.
Abstract sourceA 2027 Society of Abdominal Radiology educational exhibit is listed with Halil Ertugrul Aktas as electronic poster presenter and Gorkem Durak, Tugba Akinci D’Antonoli, and Ulas Bagci as co-authors. Acceptance and presentation status have not been verified.
Publication details
The supplied lab update identifies this work as an Educational Exhibit Abstract for the Society of Abdominal Radiology (SAR), with a scientific year of 2027. It lists Halil Ertugrul Aktas as electronic poster presenter, followed by Gorkem Durak, MD; Tugba Akinci D’Antonoli, MD; and Ulas Bagci, PhD. SAR’s official site describes educational exhibits as a submission category, but the available sources do not establish this abstract’s acceptance or presentation.
- Type: Conference
- Venue: Society of Abdominal Radiology (SAR)
- Year: 2027
- Electronic poster presenter: Halil Ertugrul Aktas
- Co-authors, in supplied order: Gorkem Durak, MD; Tugba Akinci D’Antonoli, MD; Ulas Bagci, PhD
Cite this work
Halil Ertugrul Aktas; Gorkem Durak, MD; Tugba Akinci D’Antonoli, MD; Ulas Bagci, PhD (2027). Explainable artificial intelligence in medical imaging: how to interpret, evaluate, and use artificial intelligence explanations. Society of Abdominal Radiology (SAR)
BibTeX