AI Model Predicts Indwelling Catheter Conversion in Early Neurogenic Bladder

THURSDAY, Aug. 27, 2026 (HealthDay News) -- A novel prognostic stratification model can predict indwelling catheter conversion (ICC) among patients with early-stage spinal cord injury (SCI) and neurogenic bladder (NB), according to a study published online Aug. 23 in Neurourology and Urodynamics.Jiqiang Xie, from the Fourth Military Medical University in Xi'an, China, and colleagues retrospectively analyzed 135 consecutive patients with early-stage SCI-NB admitted between November 2017 and January 2026 to develop and validate interpretable machine learning (ML) survival models for predicting indwelling catheter time (ICT). Five ML survival models were developed to estimate the probability of ICC within one year.The median ICT was 72 days in the cohort. During follow-up, 102 patients (75.56 percent) experienced ICC. The researchers found strong discriminative performance for the final SHapley Additive exPlanations (SHAP)‐informed five‐variable Random Survival Forest (RSF) model, with a validation C-index of 0.8742. For the final RSF model, time-dependent area under the receiver operating characteristic curve values were 0.9352, 0.8834, and 0.8803 at three, six, and 12 months, respectively. Five dominant early obtainable predictors were identified in the SHAP analysis: H-reflex, lower extremity motor score, urinary tract infection, time from lesion to rehabilitation facility, and spinal cord independence measure bladder score. Patients were stratified into three distinct prognostic groups using an RSF-derived prognostic output generated by the final five‐variable RSF model: favorable, intermediate, and unfavorable prognosis, with three-month cumulative ICC incidence rates of 97.30, 50.23, and 2.22 percent, respectively."Following successful external validation, this tool could support personalized bladder management strategies by implementing specific prognosis‐stratified clinical pathways," the authors write.Abstract/Full Text (subscription or payment may be required).Sign up for our weekly HealthDay newsletter

AI Model Predicts Indwelling Catheter Conversion in Early Neurogenic Bladder

Related Stories

Model May Help Identify Women at Risk for Overactive Bladder

Protocol IDs Patients Who Can Be Treated While Seated in the ED

AI Can Accurately Detect Left Ventricular Dysfunction From Electrocardiograms

Model IDs Inpatients at Risk for Hypoglycemia Within 24 Hours