Evolving visual counterfactual medical imagery explanations with cooperative co-evolution using dynamic decomposition

Authors

Luke Kelly, Edith Cowan UniversityFollow
Martin Masek, Edith Cowan UniversityFollow
Chiou Peng Lam, Edith Cowan UniversityFollow
Brandon Abela, Edith Cowan UniversityFollow
Ashu Gupta

Author Identifier

Luke Kelly: https://orcid.org/0000-0002-5194-3821

Martin Masek: https://orcid.org/0000-0001-8620-6779

Chiou Peng Lam: https://orcid.org/0000-0002-4843-9229

Brandon Abela: https://orcid.org/0000-0001-5764-9554

Document Type

Conference Proceeding

Publication Title

GECCO 2024 Companion - Proceedings of the 2024 Genetic and Evolutionary Computation Conference Companion

First Page

291

Last Page

294

Publisher

Association for Computing Machinery

School

School of Science

RAS ID

71870

Comments

Kelly, L., Masek, M., Lam, C. P., Abela, B., & Gupta, A. (2024, July). Evolving visual counterfactual medical imagery explanations with cooperative co-evolution using dynamic decomposition. In Proceedings of the Genetic and Evolutionary Computation Conference Companion (pp. 291-294). https://doi.org/10.1145/3638530.3654224

Abstract

Explainable AI (XAI) is becoming increasingly vital in healthcare, particularly in understanding decisions made by complex, black-box models in medical diagnostics. Amid discussions on balancing AI's predictive accuracy with interpretability, 'post-hoc' XAI methods are a potential solution, offering interpretability after model training without compromising diagnostic precision. Counterfactual explanations are one such 'post-hoc' XAI technique, where these explanations modify input data, like features in an X-ray image, to see how such changes affect AI image-driven diagnoses, offering insights into the model's reasoning. However, the complexity of medical image data poses challenges in generating these counterfactuals. To address this challenge, we propose a novel Cooperative Co-evolution approach with dynamic decomposition for explainability in AI for medical diagnostics by evolving visual counterfactual explanations for medical images without requiring preliminary assumptions about image decomposition. The method, evaluated using MNIST and Diabetic Foot Osteomyelitis datasets and benchmarked against a genetic algorithm, was capable of generating counterfactual explanations for medical images, showing advantages with larger images.

DOI

10.1145/3638530.3654224

Access Rights

subscription content