A prospective study of 617 pregnancies found that using unsupervised machine learning on first-trimester fetal cardiac parameters can effectively stratify risk for small-for-gestational-age (SGA) birth weight. Two key metrics — chest area z-score (P=.031) and tricuspid valve E/A ratio (P<.001) — were used to generate three risk clusters: Low (n=202, 1.2% SGA rate), Intermediate (n=217, 5.4%), and High (n=198, 14.4%) (P<.001) . The model achieved an AUC of 0.71 and demonstrated 95.5% specificity and a 99% negative predictive value for ruling out SGA — a powerful tool for early reassurance.
To read more ; Click here Could this data-driven approach redefine first-trimester risk screening and enable earlier, targeted interventions? ##Reference## Horgan, Rebecca MD; Sinkovskaya, Elena MD, PhD; Kalafat, Erkan MD, MS; Saade, George MD; Abuhamad, Alfred MD. Small-for-Gestational-Age Birth…
