Artificial intelligence and medical diagnosis: past, present and future

Area of Science:

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

Background:

• Health information technology (HIT) can reduce diagnostic error, as suggested by the NASEM report.
• Computer-based diagnostic decision support systems (CDSS) have a long history but limited clinical impact.
• Current research highlights the potential of AI-enabled CDSS.

Purpose of the Study:

• To review the historical development of AI in medical diagnosis.
• To analyze the current landscape of AI-enabled diagnostic decision support systems.
• To anticipate the future role of AI in reducing diagnostic error.

Main Methods:

• Literature review of historical and current research on AI in diagnosis.
• Analysis of the integration of AI-enabled CDSS into clinical workflows.
• Discussion of the potential impact of AI on diagnostic error reduction.

Main Results:

• AI-enabled decision support systems, when properly integrated, are expected to play an increasing role in reducing diagnostic errors.
• Despite a long history, CDSS have not significantly impacted clinical practice to date.
• AI holds significant promise for improving diagnostic accuracy.

Conclusions:

• Careful implementation of health information technology, particularly AI-enabled systems, is crucial for reducing diagnostic error.
• AI-driven diagnostic tools are poised to become integral to clinical practice.
• The future of diagnosis will likely involve advanced AI support systems.