eXplainable Artificial Intelligence for Hip Fracture Recognition

Detecting hip fractures from X-rays is a critical area where artificial intelligence can significantly reduce diagnostic errors, minimize reliance on advanced imaging techniques, and expedite the diagnostic process and subsequent surgical interventions. In this paper, we present an approach of eXpla...

Full description

Saved in:
Bibliographic Details
Main Authors: Enrique Queipo-de-Llano, Marius Ciurcau, Alejandro Paz-Olalla, Belén Díaz-Agudo, Juan A. Recio-García
Format: Article
Language:English
Published: Taylor & Francis Group 2025-12-01
Series:Applied Artificial Intelligence
Online Access:https://www.tandfonline.com/doi/10.1080/08839514.2025.2502568
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Detecting hip fractures from X-rays is a critical area where artificial intelligence can significantly reduce diagnostic errors, minimize reliance on advanced imaging techniques, and expedite the diagnostic process and subsequent surgical interventions. In this paper, we present an approach of eXplainable Artificial Intelligence, which focuses not only on the accuracy of models but also on their interpretability and the ability of users to understand and trust the decisions made by the automatic system. We present a model for the automatic classification of hip fractures in radiographs based on Convolutional Neural Networks, which is enhanced by a twin Case-Based Reasoning methodology for acquiring explanatory experiences and learning how to generate textual explanations. These findings underscore the practical benefits of incorporating explanations into medical diagnostics, paving the way for improved patient outcomes and more reliable diagnostic processes.
ISSN:0883-9514
1087-6545