Ethical layering in AI-driven polygenic risk scores-New complexities, new challenges

Area of Science:

• Genomics
• Bioinformatics
• Medical Ethics

Background:

• Polygenic risk scores (PRS) are developing for disease prevention and treatment.
• Machine learning, especially deep neural networks, is increasingly used to create PRS from health data.
• Existing ethical discussions on PRS often overlook the implications of AI integration.

Purpose of the Study:

• To address the ethical implications of AI-driven PRS.
• To highlight the need for urgent consideration of AI ethics in PRS research and clinical translation.
• To explore the complexities arising from the confluence of AI and PRS.

Main Methods:

• Comprehensive literature review on AI-driven PRS, ethical implications, and AI ethics challenges.
• Analysis of emerging complexities in fairness, trust, explainability, and regulation.
• Identification of ethical layers in AI-driven PRS development and implementation.

Main Results:

• AI integration in PRS offers potential for improved accuracy and prediction.
• Significant ethical challenges emerge concerning fairness, trust, and explainability of AI-driven PRS.
• Regulatory uncertainties and practical implementation hurdles require attention.

Conclusions:

• Proactive integration of ethical considerations is essential for AI-driven PRS research.
• Addressing ethical complexities is vital for the responsible translation of AI-driven PRS into clinical practice.
• Further research and discussion are needed to navigate the ethical landscape of AI in personalized medicine.