Artificial intelligence in clinical decision support systems for oncology.

Area of Science:

• Medical Informatics
• Artificial Intelligence in Medicine
• Oncology Decision Support

Background:

• Artificial intelligence (AI) is increasingly utilized in medical fields like image diagnosis and treatment planning.
• AI's application in tumor image-aided diagnosis demonstrates mature human-computer interaction.
• Ethical concerns and incomplete human-computer interaction limit AI's role in clinical decision-making.

Purpose of the Study:

• To summarize AI applications in clinical decision support systems (CDSS).
• To clarify the principles of AI within CDSS.
• To analyze challenges of AI in oncology decision-making and propose future solutions.

Main Methods:

• Literature review of AI applications in medical fields, focusing on CDSS.
• Analysis of existing AI-driven CDSS, including Watson for Oncology and CSCO AI.
• Identification and discussion of ethical and practical barriers to AI implementation in oncology.

Main Results:

• AI has diverse applications in medical diagnosis, treatment selection, and prognosis.
• Current AI-driven CDSS face limitations in clinical practice due to ethical and interaction issues.
• Existing AI CDSS are being promoted globally, indicating growing adoption.

Conclusions:

• AI holds significant potential for enhancing oncology clinical decision support.
• Addressing ethical considerations and improving human-computer interaction are crucial for AI adoption in CDSS.
• Future strategies are needed to fully realize AI's capabilities in oncology decision-making.