Recommendations for using artificial intelligence in clinical flow cytometry

Area of Science:

• Clinical Hematology
• Medical Informatics
• Computational Biology

Background:

• Flow cytometry is essential for diagnosing hematologic malignancies.
• Current methods rely on expert interpretation of complex data.
• Artificial intelligence (AI) offers potential advancements in data analysis and case prioritization.

Purpose of the Study:

• To review critical considerations for applying AI in clinical flow cytometry.
• To provide practical guidance for AI development and implementation.
• To suggest regulatory frameworks for AI in clinical laboratories.

Main Methods:

• Multidisciplinary stakeholder review of AI applications in flow cytometry.
• Analysis of use case identification, risk assessment, validation, and computational factors.
• Examination of current regulatory landscapes for AI in medicine.

Main Results:

• AI has the potential to improve efficiency, reduce errors, and uncover new biological insights in flow cytometry.
• Key considerations include appropriate use case selection, rigorous validation, and computational infrastructure.
• Existing regulatory frameworks need adaptation for AI in clinical settings.

Conclusions:

• AI integration in clinical flow cytometry requires careful planning and validation.
• Practical guidance and regulatory suggestions are provided for AI development and implementation.
• Ongoing updates and collaboration are necessary as AI technology in this field matures.