Body / AbdominalAI / InformaticsResearch

CT Radiomics Model Predicts Long-Term Renal Prognosis in AAV

Frontiers in digital healthSep 30

A radiomics-clinical model using renal volume, creatinine, and C3 predicted 3-year end-stage renal disease with AUC 0.93 vs 0.91 for clinical-only in 262 ANCA-associated vasculitis patients.

  • Retrospective cohort of 262 AAV patients (mean follow-up 40.8 months); 23% developed ESRD.
  • Renal volume was independently associated with ESRD (Wald p=0.004).
  • The combined model improved risk reclassification for the 3-year composite outcome (IDI p=0.018, NRI p=0.024).

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.