EVIDENCE

Measured against birth weight.

814 real pregnancies were replayed scan by scan and checked against each baby's birth weight.

58%

of the babies a 30-week scan misses, caught where the rescan Growth Intelligence asked for was done. Across the whole cohort the figure is 24%, because care did not always follow.1

15×

fewer rescans, as many small babies found: 22 rescans instead of 340, a median 4.4 weeks before birth instead of 1.32

814

pregnancies with recorded birth outcomes1

Internal validation, 814 pregnancies, October 2026.

CALLED LARGE, BORN NORMAL

One in fourteen babies the standard estimate called large were born normal size.3

Of 123 pregnancies whose last standard estimate from 32 weeks was above the 90th centile, 9 were born within the normal range. In the month before birth, the standard estimate reads 5 to 7 percent high.

When a baby is wrongly called large, the chance of a cesarean during labor more than doubles.4

Growth Intelligence reads within 1 percent.

MORE SCANS, SHARPER ANSWERS

Every scan makes the assessment more certain.

Birth weight forecast from a pregnancy's first scans: how far off it was, on average.5

Error in birth weight

Three scans

8.5%

151 pregnancies

One scan

11.0%

217 pregnancies

The scans are already in the chart.

The number drives the delivery

“For suspected macrosomia, the accuracy of estimated fetal weight using ultrasound biometry is no better than that obtained with abdominal palpation.”

ACOG Practice Bulletin 216, 2020

The estimated fetal weight decides who gets watched, who gets induced, and who gets a cesarean.6 It is least reliable at the extremes, where those decisions get made.

ACOG’s own bulletin puts ultrasound sensitivity for a baby over 4,500 g at 10 to 45 percent.7 One in four fetuses estimated at the 20th percentile is actually below the 10th.8

How weight estimation shapes clinical decisions

Estimated fetal weight is frequently the most important factor guiding clinical decisions during pregnancy, including heightened surveillance, induction of labor, and cesarean delivery (Kadji et al., Am J Obstet Gynecol 2022). Below the 10th percentile for gestational age, a fetus is classified small for gestational age and enters a surveillance pathway: Doppler studies, serial ultrasound, early delivery planning (ACOG Practice Bulletin 227, 2021). Above 4,500 grams, it triggers counseling for cesarean delivery to avoid shoulder dystocia (ACOG Practice Bulletin 216, 2020). For diabetic pregnancies, this threshold is 4,000 grams.

Why there is a detection gap

The formula used in the vast majority of clinical settings dates to 1985 and was built from 167 fetuses at a single hospital (Hadlock et al., 1984 and 1985). No general-purpose formula developed since has outperformed it on aggregate metrics (Hammami et al., Ultrasound Obstet Gynecol, 2018), but aggregate performance conceals the problem at the clinical margins, where decisions carry the highest stakes. Approximately half of all fetal growth restriction cases go undetected, even in high-income countries (McCowan et al., 2018).

METHOD

Assessed as it would have looked at the time.

Each pregnancy in the cohort is replayed one scan at a time, using only what was known at that scan. The standard estimate is read on the Hadlock 1991 chart, as reporting software prints it. Size at birth is placed on the INTERGROWTH-21st newborn standard. Every flag is then scored against the recorded birth weight.

A Retrospective runs the same method on your own deliveries.

Get GIRequest a Retrospective

The legal context: Standard of care

  1. Validation cohort of 814 pregnancies with recorded birth outcomes, enriched for growth abnormality (14 percent small and 31 percent large for gestational age at birth). Among the 499 with a full scan at 30 to 34 weeks, the standard estimate at that scan flagged 48 of 92 babies born small and 60 of 138 born large; it missed 122. Growth Intelligence projected each pregnancy to delivery, read growth rate, and asked for a rescan when the projected centile fell below the 25th or above the 75th. Of the 122 missed babies, 50 were in pregnancies it flagged that were rescanned at 35 to 38 weeks: 29 caught (58 percent). 18 were flagged and not rescanned: none caught. 54 were not flagged. Across all 122: 29 caught (24 percent), for 6 more false alarms among 407 babies who were not small and 8 more among 361 who were not large. Rescans were ordered by clinicians, not by Growth Intelligence, and rescanned pregnancies were small 1.71 times as often. Cross-validated. Internal analysis, not yet externally validated. Data on file, October 2026.
  2. Same cohort. Rescans, small babies only: in 340 pregnancies scanned at 30 to 34 weeks and again at 35 to 38 weeks, rescanning all 340 found 41 of 71 small babies, a median 1.3 weeks before birth, with 13 false alarms among 269. Growth Intelligence's rescan rule asked for 22 rescans (15 times fewer) and found 42, a median 4.4 weeks before birth, with 18 false alarms, the same as with no rescan. For large babies the rule asks for few rescans and adds one. On the 19 held-out small babies the policies did not differ. Cross-validated unless stated. Internal analysis, not yet externally validated. Data on file, October 2026.
  3. Same cohort. Of 123 pregnancies whose last standard estimate from 32 weeks was above the 90th centile, 9 were within the normal range at birth (standard estimate: Hadlock HC, AC, FL, read on the Hadlock 1991 chart, as reporting software prints it; size at birth on the INTERGROWTH-21st newborn standard). In 340 pregnancies scanned at 35 to 38 weeks, the standard estimate read 4.6 percent high on average for babies born within two weeks of the scan and 6.5 percent high at two to four weeks; Growth Intelligence's weight read 0.6 percent low and 1.0 percent high. Cross-validated. Internal analysis, not yet externally validated. Data on file, October 2026.
  4. Papaccio et al., Int J Gynecol Obstet 2021: among pregnancies at risk of macrosomia scanned at 34 to 37 weeks, cesarean delivery during labor was 19 percent when the fetus was wrongly estimated as large, against 8.7 percent when it was correctly estimated as normal size.
  5. Same cohort, 245 held-out pregnancies (217 with an eligible scan from 22 weeks). Birth weight forecast from each pregnancy's first one or three eligible scans; error is the average miss as a share of birth weight. Internal analysis, not yet externally validated. Data on file, October 2026.
  6. Kadji et al., Am J Obstet Gynecol 2022.
  7. ACOG Practice Bulletin 216, 2020.
  8. Lappen & Myers, Am J Obstet Gynecol 2017.

Not an FDA-cleared device. Clinical decision support under the CDS exemption.