Functional linear discriminant analysis: a new longitudinal approach to the assessment of embryonic growth

Functional linear discriminant analysis (FLDA) is a new growth assessment technique using serial measurements to discriminate between normal and abnormal fetal growth. We used FLDA to assess and compare growth in live pregnancies destined to miscarry with those remaining viable.

This was a prospective cohort study of women with ultrasound scans on at least two separate occasions showing live pregnancies. Serial crown-rump length (CRL), mean gestational sac diameter and mean yolk sac diameter measurements were recorded. The ability of FLDA to predict subsequent miscarriage was compared with that of a single CRL measurement.

Of 521 included pregnancies, 493 (94.6%) remained viable at 14 weeks and 28 (5.4%) miscarried. The CRL growth rate was significantly lower in those that miscarried (one-sample t-test, P = 2.638E-22). The sensitivity of FLDA in predicting miscarriage from serial CRL measurements was 60.7% and specificity was 93.1% [positive predictive value (PPV) 33.3%, negative predictive value (NPV) 97.7%]. This was significantly better for predicting miscarriage than a single CRL observation of more than 2SD below that expected (sensitivity 53.6%, specificity 72.2%, PPV 9.9%, NPV 96.5%).

FLDA discriminates between normal and abnormal growth to predict miscarriage with high specificity. FLDA predicts miscarriage better than a single observation of a small CRL.