Session:
Conference Welcome
The 2025 ERA Symposium, themed “Kidney Intelligence: Bridging AI, Math, and Medicine,” opened with remarks from Len Usvyat and Doris Fuertinger. Together, they welcomed participants to Vienna and underscored the symposium’s mission — connecting clinicians, researchers, and innovators to explore how artificial intelligence and computational modeling are redefining kidney care.
Speaker:
Len Usvyat, PhD, Head of Renal Research Institute
Doris Fuertinger, PhD, Global Lead, Computational Medicine
Session:
How AI is Reshaping Nephrology: Mayo Clinic's Experience
This session highlights how artificial intelligence is transforming kidney care at the Mayo Clinic. Dr. Wisit Cheungpasitporn presents real-world applications of AI in nephrology, emphasizing its impact on clinical practice, nephrology education, and research. The talk showcases how tools like Abridge, RecordTime, OpenEvidence, and Copilot are being integrated into patient care workflows, streamlining communication, enhancing medical education, and supporting data-driven research at Mayo Clinic. Ethical considerations and future directions for responsible, AI-enhanced kidney care will also be discussed.
Speaker:
Wisit Cheungpasitporn, MD, FACP, FASN, FAST
Professor of Medicine, Mayo Clinic, Division of Nephrology and Hypertension, Rochester, MN
Session:
AI Perspectives from MGH: Applications in Fluid Management and Dialysis
This session will describe applications of AI for fluid and electrolyte disorders across diverse patient populations. Specific recent advances about the dialysis population will be discussed. The session will involve a didactic presentation.
Speaker:
Sagar Nigwekar, MD, MMSc
Co-director, Kidney Research Center, Massachusetts General Hospital, Boston, MA
Session:
Bridging AI, SDOH, and Kidney Outcomes
In this session, Dr. Lili Chan discusses how artificial intelligence can help identify and interpret social determinants of health (SDOH) that influence kidney outcomes. Drawing on research from Mount Sinai’s diverse patient population, her team analyzed data from dialysis patients to explore how factors such as housing instability, food insecurity, and transportation barriers affect quality of life and hospitalization rates.
Dr. Chan also demonstrates how large language models (LLMs) can extract SDOH insights directly from electronic health records, offering a scalable, AI-driven approach to understanding social risk factors. Her work underscores the importance of integrating equity-focused analytics into nephrology to build more inclusive, data-informed models of care.
Speaker:
Lili Chan, MD
Associate Professor, Icahn School of Medicine at Mount Sinai Division of Nephrology, New York, NY
Session:
Reinforcement Learning to Guide Dynamic Treatment Regimens for Chronic Kidney Disease-Mineral and Bone Disorder (CKD-MBD)
This session will discuss how the challenge of interacting with medication classes, complex phenotypes, and dynamic feedback in the treatment of CKD-MBD is ideally suited for artificial intelligence methods such as reinforcement learning. Dr. Scialla will relay her team’s current experience with developing dynamic treatment algorithms with a focus on assessment for clinical relevance and safety. The potential of these models to support the achievement of customized CKD-MBD goals and to support urgently needed CKD-MBD trials provides vital context to motivate the session.
Speaker:
Julia Scialla, MD, MHS, FASN
Associate Professor of Medicine and Public Health Sciences, University of Virginia School of Medicine, Charlottesville, VA
Session:
From Math to Medicine: A Virtual Parathyroid Gland to Support Calcimimetic Therapy
In this session, we highlight the power and promise of mathematical modeling in clinical research. Beginning with core principles, we demonstrate how a physiologically grounded virtual parathyroid gland, combined with a pharmacokinetic/pharmacodynamic model of a calcimimetic, can inform and support effective calcimimetic therapy.
Speaker:
Gudrun Schappacher Tilp, PhD
Professor, FH Joanneum - University of Applied Sciences, Graz, Austria
Session:
The Reality of Deploying AI in Clinical Settings: Industry's Perspective
Dr. Francesco Bellocchio from the Renal Research Institute shares an industry-driven perspective on deploying artificial intelligence in clinical nephrology. Drawing from over a decade of experience, including the pioneering launch of the Anemia Control Model in 2013, the talk highlights how Fresenius Medical Care has built and integrated AI tools to deliver measurable value in patient care. From decision-support systems to predictive analytics and digital scribes, the session explores a diverse AI portfolio designed to optimize outcomes, support clinical decision-making, and enhance operational efficiency. Attendees will gain insights into real-world challenges and success factors in implementing AI at scale, as well as future directions involving large language models and semantic technologies.
Speaker:
Francesco Bellocchio, PhD
Senior Data Scientist for Global Operational Applied Artificial Intelligence and GenAI, Renal Research Institute, Milan, Italy
Session:
AI and the Human Element in Medicine | ERA Symposium Morning Q&A
The Morning Q&A session brought together clinical and research leaders to discuss the intersection of artificial intelligence and the human aspects of care. Moderated by Dr. Peter Kotanko, the panel explored physician perspectives on trust, collaboration, and burnout, emphasizing that AI’s greatest potential lies in enhancing—not replacing—the clinician’s role.
Panelists shared insights from their own research and deployment experiences, highlighting the importance of stakeholder engagement, ethical governance, and maintaining patient-centered care as AI continues to shape the future of nephrology.
Speaker:
Peter Kotanko, MD, FASN
Adjunct Professor, Icahn School of Medicine at Mount Sinai, New York, NY
Session:
How Large-Scale Data Collection and Machine Learning Shape Quality Management in Emergency Care
Cardiac arrest remains a leading cause of death in the Western world, with survival rates around 10%. Timely treatment, including defibrillation and cardiopulmonary resuscitation, is critical for survival. To improve outcomes, analyzing real-world data is essential, yet difficult to obtain. Beyond documentation by emergency medical services, defibrillator recordings are a key data source but are typically limited by proprietary software that allows only predefined analyses.
This presentation summarizes the results of several projects focused on analyzing defibrillator and physiological data from cardiac arrest cases. A standardized framework was developed to process and annotate data independent of proprietary tools. Using this, a novel algorithm was created to detect chest compressions from accelerometer-based feedback sensors. Additionally, these sensor data were combined with ECG recordings in a machine-learning approach to predict the patient’s circulatory state.
The presented projects highlight the potential of leveraging large-scale real-world defibrillator data to advance cardiac arrest research and support the development of data-driven treatment strategies.
This work is result of a collaboration with several colleagues from the University of Graz, the Medical University of Graz, the University Hospitals Schleswig-Holstein and Münster, and the German Society of Anaesthisiology & Intensive Care Medicine.
Speaker:
Martin Holler, PhD
Professor, Idea_Lab, University of Graz, Graz, Austria
Session:
AI Ergonomics: The Human Factor in AI-Assisted Medicine
Cognitive and computational limitations of the human mind inherently constrain diagnostic accuracy, making occasional errors unavoidable, even for experienced clinicians. Turning to peers - or increasingly, to AI systems - offers a powerful means of mitigating these shortcomings. Yet the responsibility of integrating one or more external inputs with one’s own clinical judgment remains with the decision-maker. This presentation examines how effectively humans perform this integration, revealing generally reasonable outcomes, but also substantial variability across individuals and clear potential for improvement.
Speaker:
Carlo Reverberi, MD, PhD
Professor, Università Milano - Bicocca & Milan Center for Neuroscience
Session:
How Nephrologists Can Leverage Coding & Python to Improve Patient Care
In this practical and forward-thinking session, Dr. Hendrik Dannemeyer demonstrates how clinicians can use coding to solve real-world challenges in nephrology. Facing staff shortages and scheduling inefficiencies, his team developed Outpatient Dispatch, a Python-based platform that automates patient prioritization and improves clinic operations.
Dr. Dannemeyer’s experience shows that coding is more than a technical skill—it’s a mindset for innovation. By combining digital literacy with clinical insight, nephrologists can create tailored solutions that enhance efficiency, empower teams, and deliver better outcomes for patients.
Speaker:
Hendrik Dannemeyer, MD
Medical Director, NephroCare Hamburg-Altona GmbH
Session:
The Clinician-Coder Perspective: Data Science in Vascular Access Care
In this insightful session, Dr. Amun Hofmann bridges two worlds—the operating room and the code editor—to show how clinicians can harness data science to improve vascular access outcomes. Drawing from his dual background in surgery and programming, Dr. Hofmann highlights how interdisciplinary collaboration and digital literacy are essential for the future of precision medicine.
He discusses the real challenges of integrating AI into vascular access research, including data collection, model validation, and communication between clinical and computational teams. Dr. Hofmann emphasizes that while prediction models remain limited, they reveal valuable patterns that can inform patient care. More importantly, he argues that the next generation of clinicians should focus not on becoming programmers, but on understanding how to work effectively with AI and data-driven tools.
Speaker:
Amun Hofmann, MD
Department of Vascular and Endovascular Surgery, Klinik Ottakring, Vienna, Austria
Session:
The Challenges of IT and AI Integration
In this insightful session, Dr. Amun Hofmann bridges two worlds—the operating room and the code editor—to show how clinicians can harness data science to improve vascular access outcomes. Drawing from his dual background in surgery and programming, Dr. Hofmann highlights how interdisciplinary collaboration and digital literacy are essential for the future of precision medicine.
He discusses the real challenges of integrating AI into vascular access research, including data collection, model validation, and communication between clinical and computational teams. Dr. Hofmann emphasizes that while prediction models remain limited, they reveal valuable patterns that can inform patient care. More importantly, he argues that the next generation of clinicians should focus not on becoming programmers, but on understanding how to work effectively with AI and data-driven tools.
Speaker:
Emel Hamilton, MD, MSN/INF, CNN, CGAIB
Global Leader of Clinical Systems, Fresenius Medical Care
Session:
AI, Trust, and the Human Role in Medicine
The Afternoon Q&A session closed the ERA Symposium with a thoughtful exploration of the relationship between clinicians and AI. Moderated by Dr. Lili Chan, the discussion addressed how trust, education, and feedback can shape the responsible use of AI in medicine.
Panelists debated the balance between augmentation and automation, the importance of digital literacy, and the ethical challenges of integrating AI into decision-making. Their insights underscored that while algorithms can process data, only humans can bring empathy, accountability, and clinical context to patient care.
Speaker:
Lili Chan, MD
Associate Professor, Icahn School of Medicine at Mount Sinai Division of Nephrology, New York, NY
Session:
Patient Panel - The Patient Perspective on AI in Kidney Care
The final session of the ERA Symposium brought the conversation full circle — from algorithms to individuals. Moderated by Drs. Hendrik Dannemeyer and Amun Hofmann, the patient panel offered a candid and deeply human look at how AI is perceived by those living with kidney disease.
Patients Claus Pohnitzer and Nicole Cihelna discussed data privacy, trust, communication, and the value of keeping a human connection in an increasingly digital healthcare environment. Their stories reminded clinicians and researchers alike that innovation must serve people first — empowering, not overwhelming, the patients it aims to help.
Speaker:
Hendrik Dannemeyer, MD; Amun Hofmann, MD; Klaus Pohnitzer; Nicole Cihelna
