Kidney Week

Abstract: PUB494

Eco-friendly Innovation in Kidney Transplants: Outcomes from a Collaborative Artificial Intelligence (AI) Feedback Loop

Session Information

• Publication Only

Category: Transplantation

• 2102 Transplantation: Clinical

Authors

• Balakrishnan, Suryanarayanan, Mayo Clinic Minnesota, Rochester, Minnesota, United States

Suryanarayanan Balakrishnan, MBBS

• Garcia Valencia, Oscar Alejandro, Mayo Clinic Minnesota, Rochester, Minnesota, United States

Oscar Alejandro Garcia Valencia, MD

• Abdelgadir, Yasir, Mayo Clinic Minnesota, Rochester, Minnesota, United States

Yasir Abdelgadir, MD

• Thongprayoon, Charat, Mayo Clinic Health System, Mankato, Minnesota, United States

Charat Thongprayoon, MD, FASN

• Miao, Jing, Mayo Clinic Health System, Mankato, Minnesota, United States

Jing Miao, MD, PhD, FASN

• Leeaphorn, Napat, Mayo Clinic in Florida, Jacksonville, Florida, United States

Napat Leeaphorn, MD, FASN

• Craici, Iasmina, Mayo Clinic Minnesota, Rochester, Minnesota, United States

Iasmina Craici, MD

• Cheungpasitporn, Wisit, Mayo Clinic Minnesota, Rochester, Minnesota, United States

Wisit Cheungpasitporn, MD, FASN

Background

Negative environmental impact of kidney transplant is significant. Sustainable practices are crucial for reducing the environmental burden while ensuring optimal patient outcomes. ChatGPT-4.0 developed ten key strategies, which were further enhanced through a collaborative feedback process involving three other cutting-edge AI agents: Claude 3.0 Opus, Gemini Advanced, and Meta AI with Llama-3.

Methods

In April 2024, a multi-stage feedback loop approach was conducted, involving GPT-4.0 and three other AI agents (Claude 3.0 Opus, Gemini Advanced, and Meta AI with Llama 3), utilized a series of feedback loops to refine and improve ten critical strategies for establishing sustainable transplants. The feedback loops continued until all agents were satisfied with the final set of strategies, at which point the strategies were executed.

Results

The refinement process included six rounds and 29 communications encompassing initial feedback, subsequent revisions and final approvals. This iterative review led to a robust set of strategies, significantly enhancing the initial proposals. The strategies achieved specific targets (see Fig. 1), such as 50% renewable energy use by 2025, and a comprehensive sustainability plan included transitioning to renewable energy, water conservation, digital transformation and carbon offsets.

Conclusion

Our study demonstrated the potential of AI collaboration in creating sustainable medical practices for 'green' kidney transplants which can guide healthcare organizations. This model highlights the effectiveness of collaborative AI in developing sophisticated solutions to complex problems, potentially serving as a blueprint for future sustainability initiatives.