Back
For best output, select "Paper Size" as "A4" and "Margin" as "0" or "None".
To save or print to PDF, please select Print Destination > Save as PDF, enable Background Graphics under "More Settings", then click "Save".
During the congress, E-Posters will be accessible to all participants on the congress website 24/7, as well as in the E-poster stations in the congress center.
Preparing your E-Poster
Please review the E-Poster format requirements carefully when preparing your E-Poster. Should your E-Poster not meet the mentioned requirements, it may not be displayed as described above.
E-Poster Submission Deadline
Please prepare and upload your E-Poster no later than March 14, 2026 11.59PM CET. After this date, you will no longer be able to prepare and upload your E-poster and it will not be displayed and accessible on the congress website.
Please follow the instructions below to input your abstract title.
Abstract titles should be brief and reflect the content of the abstract.
Large language models (LLMs) like ChatGPT, Claude, and DeepSeek are being used more often in medical education, clinical decision-making, and patient communication. While early research shows they can give useful medical insights, their accuracy and reliability vary depending on the specialty. So far, no study has compared these models specifically in nephrology, and especially in hemodialysis. Since this is a highly specialized area, it is important to see how well LLMs perform compared to expert opinion.This pilot study aims to:Test how accurate ChatGPT, Claude, and DeepSeek are when answering dialysis-related questions.See how well their responses fit different learner levels (trainee vs. fellow).Compare their answers with evaluations made by blinded nephrology experts.
A total of 44 questions were generated for this study. Two internal medicine trainees on their medicine rotation were each asked to formulate 11 questions while being introduced to hemodialysis. In addition, two nephrology fellows (with ≥1 year of subspecialty training) each generated 11 dialysis-related questions during clinical rounds.
Each question was first answered by a clinical tutor. The same questions was then posed to ChatGPT, Claude, or DeepSeek. Learners were not given specific instructions on prompting and were asked to obtain answers as they normally would during their clinical rotations.
For evaluation, each question had two corresponding answers (one from the clinical tutor and one from the assigned LLM). These paired responses were reviewed by board-certified nephrologists blinded to the source. Each answer was scored across five domains:Accuracy and completeness,Clarity and structure,Suitability to learner level,Clinical relevance and practicality and lastly Encouragement of critical thinking and further learning.Two nephrologists independently scored the same set of responses.
