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.
Chronic kidney disease (CKD) progression varies widely among individuals and is often inadequately predicted using baseline clinical measures alone. This study aimed to evaluate whether longitudinal modeling of estimated glomerular filtration rate (eGFR) trajectories, paired with machine learning, could enhance risk stratification and highlight subgroup vulnerability.
A retrospective cohort of 182 CKD patients was followed with eGFR measurements taken at 3-month intervals at Nagoya University Hospital. Linear regression was applied to estimate individual annual eGFR slopes. Patients exhibiting a decline greater than 5 mL/min/1.73m²/year were classified as rapid progressors. A random forest classifier was trained using baseline clinical and biochemical features to predict rapid decline. Kaplan-Meier–like curves and trajectory plots were used to visualize disease dynamics and progression heterogeneity (Figure 1).
The cohort had a mean age of 62.6 years, with high comorbidity rates: 50.5% had diabetes mellitus (DM), and 79.7% had hypertension (HTN) (Table 1). The average eGFR slope was –6.53 ± 6.14 mL/min/year, with 42% of patients classified as rapid progressors (Table 2). Among those with both DM and HTN, the mean eGFR decline was significantly steeper (–5.9 vs. –3.6 mL/min/year; Table 3). Trajectory plots revealed considerable heterogeneity in decline patterns (Figure 2), and subgroup comparison demonstrated more aggressive loss of kidney function among diabetic patients (Figure 3). The random forest classifier achieved moderate precision but limited recall, with an AUC of 0.34, highlighting the complexity of predicting progression using baseline features alone.
Trajectory-based modeling of eGFR provides a dynamic and individualized risk stratification strategy for CKD patients. Visual and statistical analyses reveal distinct progression patterns, particularly in diabetic and hypertensive subgroups. While machine learning using static variables shows modest accuracy, integrating temporal data may offer greater clinical utility in identifying high-risk patients early.
