Artificial intelligence in clinical practice: Quality and evidence.

Area of Science:

• Medical Informatics
• Artificial Intelligence
• Clinical Research

Background:

• The field of artificial intelligence (AI), particularly generative AI, is experiencing rapid growth.
• There's a significant increase in AI methodologies within scientific publications, indicating a potential
• AI bubble
• requiring critical evaluation of its clinical applications.

Purpose of the Study:

• To highlight the need for standards and guidelines in AI research, especially for generative AI.
• To emphasize the importance of evidence-based AI (IABE) for ensuring efficacy and safety in healthcare.

Main Methods:

• Review of current trends in AI application in scientific literature.
• Analysis of existing and emerging initiatives for AI guideline development (e.g., CONSORT AI, STARD AI, CHART collaborative).

Main Results:

• The emergence of generative AI necessitates new frameworks for evaluating its use in clinical settings.
• Existing guidelines focus on discriminative AI, with recent efforts addressing generative AI.

Conclusions:

• Scientific regulation and evidence-based AI (IABE) are essential for guaranteeing the efficacy and safety of AI applications in healthcare.
• Establishing clear standards is vital to maintain the quality of care amidst rapid AI advancement.