😊🙏🏻 Happy to share our newly published article in 📢💊 "Protocol-constrained AI enhances tacrolimus dosing accuracy in kidney transplant care" 🔗 Read the full open-access article here: doi.org/10.3389/frai.2026.18… Tacrolimus dosing after a kidney transplant is highly complex, individualized, and prone to variability. To help address this, our team developed TacroDose AI—a protocol-constrained large language model (LLM) designed to generate dosing recommendations that strictly align with established institutional guidelines. 💡 Key Findings: 🔹 After refining our model with structured output verification (TacroAI 2.0), we achieved a remarkable 91.7% protocol adherence. 🔹 Crucially, we reduced sentinel errors to just 1.0% and significantly improved reproducibility. 🔹 This provides a powerful framework for integrating generative AI with rule-based clinical protocols to support safe, clinician-supervised medication management in transplant care. A huge congratulations to my outstanding mentee and first author, Benjamin Bizer, and our fantastic fellows who drove this project! 👏✨ I am so incredibly grateful to collaborate with this brilliant team: Oscar A Garcia Valencia, Flora Kincses, Jose Arriola-Montenegro, Charat Thongprayoon, and Jing Miao, Wisit Cheungpasitporn. @josearriolam @OscarGarciaMD @jingmiao2019 @MayoClinicNeph @FrontiersIn Wisit Cheungpasitporn, MD, FACP, FASN, FAST Clinician-Scientist & Professor of Medicine, Mayo Clinic Leading AI Innovation in Kidney Care #KidneyTransplant #Nephrology #ArtificialIntelligence #GenerativeAI #Tacrolimus #DigitalHealth #ClinicalDecisionSupport #MedicalEducation #HealthcareAI #FutureOfMedicine #MayoClinic

😃✨ Last Day of the KDIGO Controversies Conference on Technological Advancements to Support Guideline-Informed Care (20–23 November 2025, Madrid 🇪🇸) @goKDIGO Today was an inspiring final day as we presented the refined summaries, integrated feedback across all eight questions, and continued shaping a cohesive synthesis for the conference report. We also engaged in focused discussions to map out the key research gaps and research priorities that will guide future work in technology-enabled, guideline-informed kidney care. 💡 🙏 We are grateful for the opportunity to collaborate with such an exceptional group of global colleagues whose insights and dedication continue to push the field forward. @FrancescoPesce 🤝 Looking ahead, we are excited for the continued partnerships, shared learning, and innovative projects that will emerge from this week. @MayoClinicNeph Conference link 🔗 kdigo.org/wp-content/uploads…