Evaluating Use of Generative Artificial Intelligence in Clinical Pathology Practice: Opportunities and the Way...

Area of Science:

• Medical Informatics
• Artificial Intelligence in Medicine
• Clinical Pathology Applications

Background:

• Generative artificial intelligence (GAI) is poised to transform healthcare, particularly clinical pathology (CP) workflows.
• Potential applications span education, data mining, decision support, and patient trend analysis within CP.

Purpose of the Study:

• To review current and potential use cases of GAI in clinical pathology.
• Focus on large language models (LLMs) and their application across CP subspecialties.
• To identify and address potential pitfalls associated with GAI implementation in CP.

Main Methods:

• Broad review of existing literature on GAI in healthcare.
• Analysis of GAI use-case scenarios within clinical chemistry, microbiology, hematopathology, and molecular diagnostics.
• Assessment of CP data utilization for GAI, considering reporting, quality metrics, and translational research.

Main Results:

• GAI presents opportunities for revolutionizing patient care and practitioner efficiency in CP.
• Specific GAI applications were identified for various CP subspecialties.
• Potential challenges include data bias, AI hallucinations, workflow integration, and user adoption.

Conclusions:

• GAI is a powerful tool for healthcare but requires cautious implementation due to inherent limitations.
• Human-in-the-loop models are recommended for GAI integration in CP.
• GAI can provide deeper insights into individual and population patient outcomes.