Unveiling the explainability imperative in medical AI

Overview

As AI systems become increasingly prevalent in medical applications, the need for explainable AI (XAI) has become crucial. This research investigates the critical issue of explainability in medical artificial intelligence (AI) systems. This project investigates methods for improving the interpretability and transparency of AI models used in medical diagnosis, treatment planning, and prognosis prediction. Understanding the reasoning behind AI-driven decisions is essential for building trust among healthcare professionals and ensuring patient safety.

Research activities

• Development of explainable AI models for medical applications.
• Comparative analysis of various XAI techniques.
• Integration of domain knowledge into AI model explanations.
• User studies with healthcare professionals to assess explainability.
• Ethical and legal considerations in medical AI explainability.
• Data visualization and interpretation techniques.
• Collaboration with clinicians and AI ethicists.

Outcomes

The key outcomes of the project will be to:

• develop novel XAI techniques tailored for medical AI applications
• evaluate the impact of explainability on clinician trust and decision-making
• propose guidelines for implementing XAI in medical AI systems.

Skills and experience

Experience in data science, and programming and visualisation using Python or R.

Keywords

• explainable AI (XAI)
• medical diagnosis
• AI ethics
• transparency in healthcare

Contact

Contact the supervisor for more information.