Estimated fetal weight calculator (Hadlock)
Enter the biparietal diameter, head circumference, abdominal circumference and femur length from your scan report and this tool gives you the estimated fetal weight using the Hadlock 1985 regression models — plus the error range around it, which is roughly plus or minus 15%. It works in your browser.
Estimated fetal weight calculator (Hadlock)
No sign-up · PrivateMillimetres, from the scan report. Leave blank if it is not printed — the tool will use a formula that does not need it.
Millimetres. This is the measurement the 1985 paper rated most highly, because head circumference barely changes when the head is a slightly unusual shape and the biparietal diameter does.
Millimetres. Required by every one of the four formulae — nothing can be estimated without it.
Millimetres. Also required by all four formulae.
Enter the measurements from your scan report, in millimetres. Abdominal circumference and femur length are the two the tool cannot do without; head circumference makes the estimate a little better still.
Nothing you type leaves your device. The whole calculation runs in your browser.
How this is calculated
FormulaThe Hadlock equations, in full
There is no single “Hadlock formula”. The 1985 paper published four regression models, one for each useful combination of measurements, so that a weight can still be estimated when a measurement is missing from the report. All four are implemented here, taken verbatim from Table II of Hadlock FP, Harrist RB, Sharman RS, Deter RL, Park SK, “Estimation of fetal weight with the use of head, body, and femur measurements — a prospective study”, American Journal of Obstetrics and Gynecology 1985;151(3):333–337 (PMID 3881966).
That table is headed “New regression models based on an expanded sample population (n = 276 fetuses)” and footnoted “Fetal measurements in cm; fetal weight in gm.” Every measurement goes in as centimetres and the answer comes out in grams. Scan reports print millimetres, so this calculator divides by ten before it does anything else.
- AC and FL:
log10(weight) = 1.304 + 0.05281 × AC + 0.1938 × FL − 0.004 × AC × FL - BPD, AC and FL:
log10(weight) = 1.335 − 0.0034 × AC × FL + 0.0316 × BPD + 0.0457 × AC + 0.1623 × FL - HC, AC and FL:
log10(weight) = 1.326 − 0.00326 × AC × FL + 0.0107 × HC + 0.0438 × AC + 0.158 × FL - BPD, HC, AC and FL:
log10(weight) = 1.3596 − 0.00386 × AC × FL + 0.0064 × HC + 0.00061 × BPD × AC + 0.0424 × AC + 0.174 × FL
BPD is the biparietal diameter, HC the head circumference, AC the abdominal circumference and FL the femur length. Weight in grams is 10 raised to the power of whichever expression above applies.
A worked example you can check on paper
Take a scan reporting HC 330 mm, AC 340 mm and FL 72 mm. In centimetres that is HC 33.0, AC 34.0, FL 7.2. Using the HC, AC and FL model:
- 1.326
- − 0.00326 × 34.0 × 7.2 = −0.79805
- + 0.0107 × 33.0 = +0.35310
- + 0.0438 × 34.0 = +1.48920
- + 0.158 × 7.2 = +1.13760
The terms total 3.50785, and 103.50785 = 3220 g — 3.22 kg, or about 7 lb 2 oz.
How accurate this actually is
Table III of the same paper gives the systematic and random error of each model against the 276 fetuses the equations were fitted to. The random error is the number that matters, and it is quoted as one standard deviation, as a percentage of the true weight:
| Model | Mean deviation | 1 SD (random error) | R² |
|---|---|---|---|
| AC, FL | 0.3% | 8.0% | 97.3% |
| BPD, AC, FL | 0.3% | 7.5% | 97.6% |
| HC, AC, FL | 0.0% | 7.5% | 97.6% |
| BPD, HC, AC, FL | 0.1% | 7.4% | 97.7% |
The paper also validated the earlier version of these models prospectively in a fresh group of 109 fetuses and reported one standard deviation of 7.3% to 7.7% depending on the model, summarising the method in its own abstract as “the accuracy of these models (1 SD = 7.5%)”.
One standard deviation of about 7.5% means roughly two thirds of estimates land within 7.5% of the true weight and about 19 in 20 land within 15%. On a 3220 g estimate that is a genuine range of about 2740 g to 3700 g. This calculator prints that range in the result panel rather than in a footnote, because it is the single most important thing to know about an estimated fetal weight and the thing people are least often told.
Which model this tool picks, and why
Left on “use whatever I have”, the calculator prefers the HC, AC and FL model whenever those three are available. The 1985 paper argues that head circumference is a better index of head size than the biparietal diameter “primarily because of its relative shape independence”, and concludes that the head circumference, abdominal circumference and femur length model “could be considered the best overall model”. The same model is the one the WHO/HRP Fetal Growth Calculator implements, the one the WHO Fetal Growth Charts study used to derive its own estimated weights, and the one the INTERGROWTH-21st group selected in 2020 when it published a standard for Hadlock’s estimation of fetal weight. If head circumference is missing the tool falls back to BPD + AC + FL, and then to AC + FL. Every model the entered measurements can support is shown alongside the headline figure, so the difference the choice makes is visible rather than hidden.
Verification
The coefficients above were read from the 1985 paper itself and then checked against two independent publications of the same equations: Stirnemann, Salomon and Papageorghiou print the three-parameter model in running text in “INTERGROWTH-21st standards for Hadlock’s estimation of fetal weight”, Ultrasound in Obstetrics & Gynecology 2020 (doi 10.1002/uog.22000), and the WHO/HRP Fetal Growth Calculator implements it in its own application code. Running identical biometry through this calculator and through that one produces answers that agree to better than a twentieth of a gram.
What this number is not
It is an estimate of size today. It is not a diagnosis, it is not a prediction of birth weight, and on its own it says nothing about whether a baby is growing well — growth is a trajectory and one measurement is not a line. The weight also has to be plotted against gestational age on a named growth chart before it means anything, and different charts give different centiles for the same weight.
The number, and the thing nobody tells you about it
Somewhere near the end of a growth scan, a sonographer types four measurements into a machine and it prints a weight. Two point one kilos. Four pounds fourteen. It looks like a fact, because it has decimal places and it came out of a computer.
It is not a fact. It is a regression equation from 1985 applied to four measurements of a moving target through a layer of abdominal wall, and it carries a random error of about 7.5% at one standard deviation. Which means that about one time in twenty, the real weight is more than 15% away from the printed one. On a 2.1 kg estimate, that is anywhere from 1.8 kg to 2.4 kg and nothing has gone wrong.
This is not a reason to distrust the scan. It is a reason to hold the number loosely, which is exactly how the people who ordered it hold it.
Where the equations came from
Frank Hadlock and colleagues at Baylor College of Medicine spent the early 1980s working out which combinations of fetal measurements predicted birth weight best. Their conclusion, published in the American Journal of Obstetrics and Gynecology in February 1985, was that you need at least three: something for head size, something for abdominal size, and something for length. Head plus abdomen alone — the older Shepard method — had a random error of about 10%. Adding the femur brought it down to about 7.5%.
Forty years on, those equations are still the default in most ultrasound machines and still the ones a systematic review picks out as the most accurate available. That is unusual longevity for a clinical formula, and it says something about how hard the problem is.
Why your report might disagree with this page
Because there are four Hadlock equations and your machine used one of them. They do not agree with each other. On a typical set of term measurements, the four models here span a range of 60 or 70 grams — not much, but enough to notice if you are comparing a number on a screen with a number on a piece of paper.
Other things that move the number: whether the abdominal circumference was measured in a clean plane, where the calipers landed, the baby’s position, how much fluid there was to see through, and how far along you are. Estimates get less accurate as babies get bigger, which is inconvenient, because late pregnancy is when the estimate is most likely to be used for a decision.
What happens with the number
On its own, very little. An estimated fetal weight only becomes meaningful once it is plotted against gestational age on a growth chart, and even then the chart matters as much as the weight. UK guidance uses an estimated weight below the 10th centile as a threshold for closer surveillance and below the 3rd as severe, but it also says plainly that serial measurements — the direction of travel across two or more scans, at least three weeks apart — tell you far more than any single estimate.
That is why a single scan rarely changes anything on its own, and why being asked back in three weeks is usually a sign that the system is working rather than that something is wrong.
If the number has frightened you
Take the range with you, not just the point estimate. “The scan said 2.1 kg” and “the scan said somewhere between 1.8 and 2.4 kg” are the same sentence, but the second one is the true one and it is much easier to think next to.
Then ask your midwife or the team who scanned you what they are watching for and when they want to see you again. They have the fluid volume, the Dopplers, your notes and the last scan, none of which fit into a calculator. Asking is never an overreaction, and nobody has ever minded.
Sources
- Estimation of fetal weight with the use of head, body, and femur measurements — a prospective study. Am J Obstet Gynecol 1985;151(3):333–337 — Hadlock FP, Harrist RB, Sharman RS, Deter RL, Park SK (PubMed, PMID 3881966), accessed
- INTERGROWTH-21st standards for Hadlock’s estimation of fetal weight. Ultrasound Obstet Gynecol 2020 — Stirnemann J, Salomon LJ, Papageorghiou AT (Wiley), accessed
- Fetal Growth Calculator — WHO/HRP and the Special Programme of Research in Human Reproduction (srhr.org), accessed
- The accuracy of ultrasound estimation of fetal weight in comparison to birth weight: a systematic review. Ultrasound 2018;26(1):32–41 — Milner J, Arezina J (PubMed, PMID 29456580), accessed
- Small-for-Gestational-Age Fetus and a Growth Restricted Fetus, Investigation and Care (Green-top Guideline No. 31) — Royal College of Obstetricians and Gynaecologists, accessed