Urologist Hsin-Hsiao (Scott) Wang and his team have developed a pioneering machine-learning model to predict dilating vesicoureteral reflux (VUR) in infants based on early postnatal ultrasound features. While traditional ultrasound is effective for detecting hydronephrosis, it falls short in diagnosing VUR, which is better assessed with a more invasive and expensive voiding cystourethrogram (VCUG). Their model, tested on data from 280 infants, achieved a strong predictive performance with an area under the curve of 0.81. The tool aims to help clinicians determine which patients are more likely to have VUR and thus benefit from VCUG, potentially reducing unnecessary testing. Wang’s team is working on an enhanced model using a larger dataset to further refine predictions and improve patient management.
MACHINE LEARNING ALGORITHM COULD OFFER UROLOGISTS A ‘CRYSTAL BALL’ FOR PREDICTING VUR
No results found.
