Applications of Explainable Artificial Intelligence in Diagnosis and Surgery

Area of Science:

• Medical Artificial Intelligence
• Explainable Artificial Intelligence (XAI)
• Clinical Decision Support

Background:

• Artificial intelligence (AI) shows potential in medicine but faces challenges with explainability.
• The
• black-box
• nature of many AI models hinders clinical adoption.
• Explainable Artificial Intelligence (XAI) aims to provide transparency and interpretability in AI systems.

Purpose of the Study:

• To review recent trends in medical diagnosis and surgical applications utilizing XAI.
• To analyze XAI methods, challenges, and future directions in medical AI.
• To provide a reference for developing medical XAI applications.

Main Methods:

• Conducted a literature survey of articles published between 2019 and 2021.
• Searched major academic databases: PubMed, IEEE Xplore, ACM, and Google Scholar.
• Extracted and analyzed relevant information from selected studies, including an experimental showcase on breast cancer diagnosis.

Main Results:

• XAI offers both decision-making capabilities and model explanations, overcoming limitations of traditional AI.
• Identified various XAI methods applied in medical diagnosis and surgical contexts.
• An experimental showcase demonstrated XAI's application in breast cancer diagnosis.

Conclusions:

• Medical XAI is a rapidly advancing and promising research field.
• XAI facilitates the integration of AI into clinical practice by enhancing trust and understanding.
• Further research is needed to address current challenges and explore future directions in medical XAI.