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.
For hemodialysis patients, the optimization of body fluid volume is highly critical, and the ultrafiltration (fluid removal) volume is the most fundamental and important management parameter in dialysis care. However, specialists do not always set target weights for each patient in dialysis facilities. Furthermore, if hypotension could be predicted in advance during fluid removal after setting a target weight, treatment could be safely performed by adjusting the amount of fluid removal at that time. These predictions have largely relied on the knowledge and experience of specialized physicians and staff. We developed the medical program to get optimized ultrafiltration volume using artificial intelligence (AI).
This AI has been trained on 400,000 instances of dialysis medical data prescribed by specialists．AI has been developed by acquiring patient data from 16 domestic facilities in Japan, totaling approximately 3,000 cases (800,000 dialysis sessions), which accounts for about of the hemodialysis patients．
For the clinical performance trial, cases of test patient data have been acquired from eight facilities spanning Tohoku, Kanto, Chubu, and Western Japan to avoid regional bias and verify the AI's performance. The AI references the patient's data from the past five dialysis sessions to predict ultrafiltration volume and other dialysis conditions. The current day's medical data such as current day's before HD body weight, dry weight, time removing body fluid volume are used for prediction.
.
The average absolute error (MAE) of the AI prediction relative to the target ultrafiltration volume prescribed by specialists was approximately 120mL on average, confirming its generalization performance. The preliminary results of the clinical performance trial show that the accuracy rate of 92.2%. There were no bias between sex, regions,facilities. The error in predicting the amount of fluid removal in the next hour from a certain time point was 20 ml per hour. The ability to predict hypotension was 0.90 in precision and 0.73 in recall, and the system was able to accurately predict the occurrence of hypotension 30 minutes later on average.
Based on these results, the developed product aims to utilize AI, which has been trained on a large volume of specialist data, to predict specific ultrafiltration plans based on individual patient data, thereby contributing to the efficiency of the medical setting while maintaining the quality of medical care.
