Wcn26-2539 Detect-Pd: Ml Models to Estimate Peritoneal Transporter Status and Dialysis Adequacy: Pilot Results From a Prospective Cohort

Key Result

Machine learning models using routine peritoneal dialysis outflow and spot urine/blood panels are being evaluated to estimate peritoneal membrane transport status and dialysis adequacy.

Key Points

• The research aims to assess if AI models can estimate peritoneal transport and dialysis adequacy using simple measurements.
• Developed AI models using data from PD outflow and urine/blood panels.
• Conducted a prospective cohort design to gather relevant data.
• Evaluated the models' ability to estimate PET and Kt/V.
• AI models showed promise in estimating PET and dialysis adequacy.
• Potential to reduce the time and labor needed for traditional testing.
• Findings suggest simpler methods may still provide accurate information.