A developmental hip screen run end to end on the linear array of a handheld probe — one coronal image, three construction lines laid on the frozen frame, and two angles read against the table Graf published — with every step timestamped, the settings read off the device screen, and the whole recording transcribed.
Acerca de la Dra. Mariz
I am Dr. Fernando Mariz, a gynecology and pelvic surgery physician practicing in New York City. Before my medical career, I served in the U.S. Marine Corps, where I developed the discipline, focus, and steadiness that continue to shape the way I care for patients today. At Maiden Lane Medical, my work covers women’s health, preventive care, sonography, pelvic pain, abnormal uterine bleeding, and minimally invasive gynecologic procedures.
Acerca del Dr. Ávila
I am Dr. Jailyn Avila, an emergency medicine physician, educator, and digital health innovator based in Southern California. My clinical work is rooted in emergency care, while my teaching focuses on point-of-care ultrasound, residency education, and practical training for physicians. Through my roles with Emergent Medical Associates, UHS SoCal MEC Emergency Medicine Residency, Core Ultrasound, and FemInEM, I work at the intersection of patient care, medical education, and accessible digital learning.
Respuesta rápida
An infant hip ultrasound answers one question — is the socket deep enough to hold the head — and the Graf method answers it from a single coronal image with two angles. You lay a baseline down the straight lateral wall of the ilium, drop a second line from it out to the bony rim of the acetabulum, and the angle between them is alpha: 60° or more is a mature Type I hip. A third line runs from the bony rim through the centre of the cartilaginous labrum, and the angle that makes with the same baseline is beta, which should come in under 55°. The hip in this recording reads 71.6° and 48.6°, and the whole measurement takes about forty seconds once the image is frozen. My advice on the order of work is the part people get wrong: do not reach for the calipers until the frozen frame shows a straight iliac line, a visible labral tip and the transition into the triradiate cartilage, because angles measured off a plane that is not the standard plane are precise and meaningless. Jump to 1:31 for the landmarks and 2:07 for the numbers.
Looking for a device rather than the exam? The pediatrics hub covers the models used for infant and paediatric work, and the D3Ultra product page carries the full specification and current price.
The construction
One image, three lines, two numbers
Everything this exam produces comes off one frozen frame. Three lines go on it in a fixed order, two of them cross the first, and the two angles that result are read against a table published four decades ago and still in use. Click any line to jump the video to the moment it is drawn.
01BaselineAlong the straight lateral wall of the iliumDrawn first and measured never. Both angles are read against it, so every degree that follows inherits the error in this one line.Drawn at 1:4402Bony roof lineLower limb of the ilium out to the bony rimIts angle to the baseline is alpha. Alpha is a depth measurement in disguise: how far the bony socket has actually come round the head.Drawn at 1:5703Cartilage roof lineBony rim through the centre of the labrumIts angle to the baseline is beta. Beta reads the cartilage the bone has not reached yet, and it rises as the roof is pushed up and out.Drawn at 2:07
Where this hip lands on the Graf table, both axes drawn to scale
Alpha · the bony roofIII / IVDislocatedIIcCriticalIIa / IIbImmature to 3 months, delayed afterIMature43506071.6°this hipBeta · the cartilage roofUnder 55Cartilage roof where it should be55 to 77Type II territoryOver 77Decentring557748.6°this hip
Alpha bands after Omeroğlu’s tabulation of Graf’s own classification; the beta threshold of 55° after Kang and Koo. Two details the bands cannot show. The 50–59° band is the same number with two different names and two different meanings — Type IIa, physiological immaturity, at three months or younger, and Type IIb, a delay of ossification, after that — so an alpha angle without an age attached is not yet a reading. And Graf’s own table splits the mature Type I at beta 55° into Ia and Ib rather than calling everything above it abnormal, which is why the page states 55° as a threshold to act on and not as a wall.
The societies are more relaxed about the angles than this band might suggest. The ACR–AIUM–SPR–SRU practice parameter defines the examination by its two orthogonal planes — a coronal view in the standard plane at rest and a transverse view of the flexed hip with and without stress — and treats measuring the alpha angle as optional validation of what the coronal view already shows. Read the numbers as a way of making the impression checkable and repeatable between operators, which is exactly what they are good for.
Timestamped
Key moments, with the settings
The third column carries what the device panel actually reported at that moment. The useful thing about this recording is how little of it there is: one depth, one frequency, one gain, set before the probe went down and never touched again. The first three rows are argued to camera against reference material, so no panel is on screen for them.
| Tiempo | What is on screen | Array, depth, frequency, output |
|---|---|---|
| 0:15 | What DDH is, and the two tests that look for itarguedThe shape problem stated plainly: a socket that forms too shallow lets the femoral head slide out. Then the clinical screen — Barlow to push the hip out, Ortolani to bring it back, a felt clunk either way — and the verdict on it, that the pair is specific but not sensitive. | No device panel on screen — argued to camera |
| 0:47 | The number that justifies scanning at allarguedA PubMed abstract is put up full-frame with one line highlighted: against ultrasonography as the reference standard, clinical examination came out at 28.1% sensitivity and 94.5% specificity. That single figure is the whole argument for imaging a hip that examined normally. | Reference screenshot — no live imaging |
| 1:00 | What the Graf method actually asks of youarguedOne image in the coronal plane, two angles measured on it, both compared to a published table. A figure and an infant-scanning photograph are credited on screen to a pediatric hip and spine review. The claim being made is that this takes under a minute. | Reference figure — no live imaging |
| 1:19 | Acquiring the coronal imageThe acquisition itself, with the leg held at roughly 90 degrees. One pass, then freeze. The tablet is at working distance rather than to camera, so the parameter panel is not readable through this window. | Linear array in use · panel not legible at this distance |
| 1:31 | The four landmarks, named on the frozen frameIlium, labrum, the bony roof of the acetabulum, femoral head — pointed out in that order on the frozen image. Those are the structures the whole construction hangs off, and the frame is held still long enough to find each one. | FREEZE · D 60 mm · F H10.0 MHz · GN 105 dB · DR 80 · MI 0.7 · TIS 0.1 |
| 1:44 | The baseline goes down firstA line laid along the straight lateral wall of the ilium, and an explicit note that no length is being measured — the tool reports 33.35 mm and the number is ignored. This line exists only to give both angles something to be measured from. | FREEZE · Meas ▸ LENGTH · V 3.6.78 · cine 45/45 |
| 1:57 | Alpha, then the start of betaThe bony roof line is dropped from the baseline vertex and alpha lands at 71.6°. The second vertex is then planted for beta and the readout opens at 13.0°, which is the caliper’s starting position rather than a measurement. | FREEZE · ANGLE tool active · both vertices on the lower limb of the ilium |
| 2:07 | The numbers, and the table they are read againstThe beta line is walked out towards the labrum and the readout passes 67.6°, 14.8° and 80.8° before settling at 48.6°. The cut-offs are then stated out loud: alpha above 60, or above 50 under three months of age, and beta below 55. This hip clears both, and the exam ends there. | FREEZE · LENGTH 33.35 mm · ANGLE 71.6° · ANGLE 48.6° |
Settings transcribed from the device interface visible in the recording: D is depth, F is transmit frequency with harmonics on, GN is gain, DR dynamic range, MI and TIS the acoustic output indices. The software title bar shows only the version, so the preset in use is not stated here. Timestamps are on the YouTube clock.
Reading the images
What this scan shows
The coronal view of an infant hip is a corner. The straight lateral wall of the ilium runs down the image as a bright vertical line, it turns at the lower limb of the ilium, and the bony roof of the acetabulum runs out from that turn towards the surface. Sitting in the corner is the femoral head, still largely cartilage at this age and therefore grey and speckled rather than white and shadowing. The labrum caps the roof as a small triangle at the far edge. Those four structures are the exam, and they are what the presenter names on the frozen frame at 1:31.
A shallow hip does not announce itself; it simply looks a little too open at that corner, which is precisely why the method converts the impression into two numbers. The plane comes first. A standard-plane image shows a straight iliac line, the tip of the labrum and the transition from the ilium into the triradiate cartilage together in one frame, and if any of the three is missing the angles that follow are measuring the wrong section of a three-dimensional socket. Get the plane, freeze, and the rest is geometry.
Alpha and beta are two views of the same failure. Alpha falls when the bony roof has not come far enough round the head; beta rises when the cartilage roof, having nothing solid under it, is pushed up and out by a head sitting too shallow. That is why a hip in trouble usually moves both numbers in opposite directions at once, and why a single number that sits close to its cut-off is worth re-measuring on a fresh image rather than reported.
Verbatim
Full transcript
Transcript — 3:06, 9 passages
Transcribed from the recording and edited for readability; square brackets mark an editorial clarification, and every timestamp jumps the video.
What DDH is
0:00My name is Brandon Ramakko, and today we’re going to be using a Suresult D3 handheld ultrasound unit to evaluate an infant’s hip for hip dysplasia, or developmental dysplasia of the hip, DDH. Now, what this is: the hip joint could be sort of malformed; the joint can be a little too shallow, making it easy for the hip to dislocate.
Ortolani and Barlow, and why they miss
0:21And the clinical test for this is called the Ortolani and Barlow test. There’s two tests: basically you dislocate the hip and relocate the hip, and if you can do it just as easy as that, and you clunk and clunk, well then you know that you have hip dysplasia. So it’s a very specific test, but those tests are not very sensitive — and in fact that combination of tests may miss two-thirds of hip dysplasia cases.
Why ultrasound, and the Graf method
0:51So what’s considered the best way of evaluating the hip for hip dysplasia is ultrasound. So wouldn’t it be great if we could, you know, with a cheap ultrasound unit, evaluate the hip in less than a minute and have our answer? And that’s what we can do. We’ll be using this and something called the Graf method. We basically take one image in the coronal plane, we measure two angles, and we compare those angles to a chart, and then we have our answer. And it’s as easy as that.
Acquiring the image
1:19I have a baby here, and we’re going to take a coronal plane image with the leg at about 90 degrees. And here we go.
Landmarks, lines and angles
1:31And that’s the image we want. We have the ilium here, we have the labrum here, we have the roof of the acetabulum here, we have our femoral head here.
1:44And I’m actually going to start with putting down a line. I don’t actually want to measure any lengths here, but this is basically the line we need to measure from for our angles.
1:57And now I have two angles I need to get: our alpha and our beta. So there’s our alpha, and we need a second angle for our beta.
The numbers
2:07Our alpha in this case is 71.6, and that needs to be larger than 60 — [larger than 60] is normal. If they’re less than three months old, then it needs to be larger than 50. And for our beta angle, we want our beta angle to be less than 55. So for my baby: it’s larger than 60, because it’s 71.6, and it’s less than 55, because it’s 48.6. So my baby is fine, which is great. And that’s it. It’s as simple as that.
2:43[Sponsor segment omitted, 2:43–3:06.]
Hardware
Device and settings
One array, one setting, the whole exam. The left column is the published specification; the right column is what the interface actually reported while this hip was being scanned, and the right column is the one to copy if you are reproducing it — with two changes I would make. An infant acetabulum sits inside the top 30 mm, so a 60 mm field spends half its pixels on tissue nobody reads; drop to 40 mm and the corner gets twice the screen. And 10 MHz is the ceiling on this linear array, which is enough for the iliac line and the bony rim but thin at the labral tip, the one structure beta depends on. A general-purpose head is not a good fit if infant hips are a weekly fixture rather than an occasional one — that is a dedicated high-frequency linear purchase.
Published specification
Suresult D3Ultra
One head, three arrays — $2,976
- Lineal7.5 / 10 MHz · 20 / 40 / 60 / 100 mm · 40 mm
- Convexo3.2 / 5.0 MHz · 90–300 mm · 45°
- Por fases3.2 / 5.0 MHz · 90–300 mm · 60°
- ImágenesB, M, Colour, Power and PW Doppler
- Beamformer192 elements · 64 channels · 256 grey levels
- RangesGain 30–105 dB · dynamic range 40–110
- MediciónLength, area, circumference, angle, trace
- Potencia2,800 mAh · 2 h scanning · wireless charging
- Handset156 × 65 × 20 mm · 263 g
- Connects toiOS, Android, Windows · dual-band Wi-Fi
Read off the screen in this recording
- ArrayLinear · rectangular field, unchanged throughout
- ProfundidadD 60 mm · deeper than an infant acetabulum asks for
- FrecuenciaF H10.0 MHz, harmonic · the ceiling on this array
- GaneGN 105 dB · wide open
- Dynamic rangeDR 80 · mid-scale of the published span
- ProcessingENH 0 · Compound OFF · no speckle reduction
- OutputMI 0.7 · TIS 0.1
- ModeB only · no M-mode, no Doppler on this exam
- SoftwareV 3.6.78 · cine buffer 45/45 on freeze
- Tools usedLENGTH for the baseline, then ANGLE twice
Same probe, other windows
Related scans
5:27Median nerve at the wrist, measured the same wayThe other exam on this device that lives or dies on one number taken off a frozen frame — anisotropy to find the nerve, then a cross-sectional area.Read the walkthrough →
6:42Rib fracture: following a cortical lineBone surface work on the same linear array, where the whole skill is knowing exactly what an unbroken bright rail looks like at 10 MHz.Read the walkthrough →
14:09eFAST trauma exam on the same probeThe long one. Pericardium, both upper quadrants, the pelvis and both lungs, scanned in a single pass with the convex and phased arrays.Read the walkthrough →All the clinical clips, filterable by anatomy and probe, are in the Scan Library.
Asked on this search
Infant hip questions
At what age is an infant hip ultrasound done?
From about six weeks of age. The ACR–AIUM–SPR–SRU practice parameter states that because of physiologic laxity, hip sonography is usually not performed on patients younger than 6 weeks unless an abnormal physical finding calls for it, and that radiography is preferable from 6 months of age onward. The infant scanned in the recording above is five months old, inside that window at both ends. My own rule is to treat a very early scan as a question rather than an answer: laxity that would have resolved on its own reads as instability, and a separate 2022 review puts the false-positive risk specifically in the first four weeks.
Source: Practice parameter for the ultrasound examination for detection and assessment of DDH. Accessed September 10, 2026.
What is a normal alpha angle on an infant hip ultrasound?
Sixty degrees or more. In the Graf classification an alpha angle of 60° or above is a Type I mature hip. Between 50° and 59° the reading depends entirely on age: at three months or younger it is Type IIa, physiological immaturity; the same 50–59° after three months is Type IIb, a delay of ossification. Below that, 43–49° is Type IIc and anything under 43° is Type III or IV. The hip measured in the recording reads 71.6°, which sits well inside Type I.
Source: Use of ultrasonography in developmental dysplasia of the hip. Accessed September 10, 2026.
What does the beta angle add that alpha does not?
Alpha reads the bone; beta reads the cartilage the bone has not reached yet. Both are measured from the same baseline along the straight lateral wall of the ilium. Alpha is the angle out to the bony rim of the acetabulum, so it describes how deep the bony socket has become. Beta is the angle out through the centre of the cartilaginous labrum, so it describes how far the cartilage roof has been pushed up and outward by a head that is not seated deeply enough. Kang and Koo put the abnormal threshold at greater than 55°. The hip in the recording reads 48.6°.
Source: Ultrasonography of the pediatric hip and spine. Accessed September 10, 2026.
How much does the clinical examination miss on its own?
Enough that a normal Barlow and Ortolani is not an all-clear. Barlow dislocates the hip, Ortolani relocates it, and a felt clunk on either is the positive finding — a click is not. In a series of 5,701 newborns screened both ways, with Graf-method ultrasound as the reference standard, clinical examination came out 28.1% sensitive and 94.5% specific: 240 of the 334 hips that ultrasound graded Graf IIb or worse were read as normal on examination. A separate review puts the physical exam at 37% sensitive for frank dislocation. Different studies and different denominators, and they point the same way — the examination is good at confirming and poor at excluding.
Source: Comparing results of clinical versus ultrasonographic examination in DDH. Accessed September 10, 2026.
Can this scan be done on a general-purpose handheld probe?
It is, in the recording above, on the linear array of a D3Ultra at 10 MHz, and all three standard-plane landmarks stay legible. Two things I would set differently. The depth is left at 60 mm when the acetabulum sits inside the top 30, so half the field is spent on tissue nobody is reading and the anatomy is drawn smaller than it needs to be. And 10 MHz is the ceiling on that array: a dedicated 12 to 18 MHz linear head resolves the tip of the labrum more crisply, and the labral tip is the structure the beta angle hangs on. Neither point stops you taking the angles. Both make them easier to defend.
Source: Página del producto Suresult D3Ultra. Accessed September 10, 2026.
Provenance
Fuentes
- Use of ultrasonography in developmental dysplasia of the hipOmeroğlu, H. Journal of Children’s Orthopaedics, 2014;8(2):105–113. Table 1 tabulates the hip types of the Graf method directly from Graf’s own 2006 text, and is the source of the alpha bands and the Type IIa / IIb age split used on this page. Accessed September 10, 2026.
- Ultrasonography of the pediatric hip and spineKang, Y. R. and Koo, J. Ultrasonography, 2017;36(3):239–251. The review shown on screen in the recording at 1:00. Source of the three standard-plane landmarks, of the definitions of the baseline, bony roof line and cartilage roof line, and of the beta threshold of 55° quoted on this page. Accessed September 10, 2026.
- Practice parameter for the ultrasound examination for detection and assessment of DDHACR–AIUM–SPR–SRU, revised 2023 (Resolution 34). Source of the six-week timing, of the preference for radiography from six months, and of the standard-plane definition. It also treats angle measurement as optional validation rather than as the examination itself. Accessed September 10, 2026.
- Comparing results of clinical versus ultrasonographic examination in DDHArti, H., Mehdinasab, S. A. and Arti, S. Journal of Research in Medical Sciences, 2013;18(12):1051–1055. The abstract displayed full-frame in the recording at 0:47. 5,701 newborns and 11,402 hips; clinical examination 28.1% sensitive and 94.5% specific against Graf-method ultrasound. Accessed September 10, 2026.
- Common ultrasound applications for pediatric musculoskeletal conditionsde Borja, C., Watkins, R. and Woolridge, T. Current Reviews in Musculoskeletal Medicine, 2022;15(6):447–455. Source of the 37% sensitivity of the physical examination for hip dislocation and of the first-four-weeks false-positive caution. Accessed September 10, 2026.
- Current evidence about developmental dysplasia of the hip in pregnancySimionescu, A. A. et al. Medicina, 2021;57(7):655. Source of the Barlow-dislocates / Ortolani-relocates definitions used on this page, and of the distinction between a felt clunk and a click. Accessed September 10, 2026.
- Página del producto Suresult D3UltraThe published attribute table: array frequencies and their depth ranges, aperture and scan angles, element and channel counts, grey levels, gain and dynamic-range spans, battery, dimensions and weight. Accessed September 10, 2026.
- Dr. Brandon RamakkoThe presenting clinician’s own site, and the basis for how his role is described here: a neuromusculoskeletal medicine specialist who uses diagnostic ultrasound routinely and teaches musculoskeletal point-of-care scanning. Accessed September 10, 2026.
- Suresult channel — the recording on this page“POCUS for DDH (Graf Method) Using Suresult D3Ultra”, published 6 October 2025, 3:06. The transcript, the key moments and every scan frame on this page come from it. Accessed September 10, 2026.
Which array does your paediatric work actually need?
An infant hip wants a high-frequency linear head and barely any depth. The abdominal and chest questions that arrive in the same clinic want the opposite. A three-in-one head answers both from one device; a dedicated pair answers both better, for more money. Tell us the setting and the case mix and you will get a direct recommendation in one conversation — one probe, two, or neither.
