Cephalometric Tracing: A Step-by-Step Guide

Cephalometric tracing means locating landmarks on a lateral cephalometric radiograph, joining them into lines and planes, and measuring the relationships between the cranial base, the jaws, the teeth and the profile. It supports the diagnosis, the treatment plan and the follow-up of a patient. This guide walks through the whole procedure in practical order: checking the cephalometric x-ray, calibrating it, placing the 15 landmarks used for the Steiner and Tweed analyses, drawing the reference planes, measuring, and avoiding the most common errors. It is written for students, residents and dentists who want a solid method, without unnecessary jargon.

What cephalometric tracing is for

Together with the panoramic radiograph, the photographs and the study models, the lateral cephalogram is part of the basic orthodontic records. It is taken at the start of treatment and at the end. The tracing answers three questions:

• diagnosis: is the discrepancy skeletal, dentoalveolar or both? In which dimension, sagittal or vertical? Is the maxilla or the mandible involved?
• planning: where should the incisors be at the end of treatment? How much space is needed? Should extractions be discussed?
• follow-up: by comparing or superimposing two tracings, you can separate what comes from growth from what comes from treatment.

Keep its limits in mind from the start. A tracing helps you decide; it never replaces the clinical examination. No single value is meaningful on its own: only measurements read together, and checked against the patient’s face, make sense. The norms come from specific populations, to which your patient does not necessarily belong. And a two-dimensional image is the projection of a three-dimensional object, with its superimpositions and distortions.

Before you trace: the lateral cephalometric radiograph

A cephalometric analysis is only as good as the radiograph it is traced on. Before placing a single landmark, check how the image was taken.

Standard conditions

• Cephalostat and Frankfort horizontal plane: the cephalostat holds the head still, orients it to the Frankfort plane and keeps the distance between the patient and the sensor constant. This standardization is what makes two successive radiographs comparable.
• Teeth in maximum intercuspation: this is the usual position for the profile. In a Class III, an additional radiograph in centric relation is recommended to quantify a possible anterior functional shift.
• Lips at rest: this is the usual instruction for assessing the profile, since tense lips distort it. Courses also describe lips-together or open-mouth views, depending on the question asked: note which one was used.
• Calibration ruler: a graduated ruler, often placed on the nose, lets you convert distances on the image into millimeters.

Magnification

The x-rays come from a point source, so the image is always slightly enlarged. The farther the source and the closer the sensor to the patient, the smaller the magnification. Courses give about 7% for a source at 1.5 m and about 1.9% at 5 m.

Two practical consequences:

• angles do not depend on the scale, but distances in millimeters do. Without calibration, U1-NA and L1-NB in millimeters or the Wits appraisal cannot be compared with the norms;
• to compare two cephalograms of the same patient, you need the same machine and the same distances.

Image quality

A good radiograph is sharp, with contrast on the bone as well as on the soft-tissue profile, and the ruler is visible. Bilateral structures (ear canals, orbits, mandibular borders) often appear doubled; a marked double image should prompt you to check the head position, or even look for an asymmetry. Finally, every radiograph must be justified: the expected benefit must outweigh the risk of the exposure.

The tracing procedure, step by step

Whether you trace on acetate or on screen, the method is the same. Classic manual tracing is done on a light box, with the profile facing right, an acetate tracing sheet and a fine pencil.

1. Check the radiograph: identity, date, position (maximum intercuspation, Frankfort horizontal, lips at rest), sharpness, ruler visible.
2. Read the image qualitatively before measuring anything: shape of the sella turcica, cervical vertebrae, cranial vault, airway, tongue position, signs of mandibular rotation. An unusual image calls for a specialist opinion, whatever the measurements.
3. Calibrate: measure the ruler on the image to convert distances into millimeters.
4. Trace the structures: soft-tissue profile, cranial base, orbit, nasal bones, pterygomaxillary fissure, ear canal, maxilla, mandible (condyle, ramus, symphysis), incisors and first molars.
5. Place the landmarks: cranial base first (S, N), then Frankfort (Po, Or), the maxilla (A), the mandible (B, Me, Go, Gn) and the incisors (incisal edges, apices, labial surfaces).
6. Draw the lines and planes: S-N, N-A, N-B, Go-Gn, Frankfort, mandibular plane, occlusal plane, incisor long axes.
7. Measure the angles, then the distances (only after calibration).
8. Check consistency: FMA + IMPA + FMIA = 180°; do FMA and SN-GoGn tell the same story? Do ANB and the Wits point the same way?
9. Interpret in order: sagittal, vertical, dentoalveolar compensations, profile, then the target incisor position.
10. Go back to the clinical findings: profile, smile, occlusion, functions. If the tracing contradicts the examination, recheck the landmarks before concluding.

The 15 cephalometric landmarks and their pitfalls

Landmarks are the foundation of the tracing. Some are anatomical (identified directly on the radiograph), others are constructed (obtained from lines). The table below lists the 15 landmarks of our free online cephalometric analysis tool, which are enough for the Steiner and Tweed analyses.

Two definitions vary between sources, and it is worth knowing:

• Gnathion is either constructed (the bisector of the angle between N-Pog and the mandibular plane crosses the symphysis) or defined more simply as the point midway between Pogonion and Menton;
• Gonion is most often constructed with a bisector in courses; some tools, including ours, define it as the most posterior and inferior point of the angle. The resulting positions can differ slightly: compare measurements obtained with the same definition.

Other analyses add Pogonion (most anterior point of the symphysis), ANS and PNS (the ends of the bony palate), point D (center of the symphysis, for SND), Articulare (for Tweed’s posterior facial height), Basion or point Xi, as well as soft-tissue profile points for the esthetic lines.

Reference lines and planes

Landmarks define lines and planes. Each analysis has its own references: the same "divergence" measured on two different planes does not give the same value.

• S-N: the anterior cranial base, and Steiner’s reference. It is considered stable after about age 8, but its inclination varies from one patient to another.
• Frankfort plane: through Porion and Orbitale. It is Tweed’s reference, and it depends on two fragile landmarks.
• Mandibular plane: Steiner uses Go-Gn; Tweed uses a plane tangent to the lower border of the mandible, through Menton. The values are therefore not interchangeable. Our free tool draws Tweed’s mandibular plane through Go-Me, which approximates it.
• Occlusal plane: from the middle of the first molar overlap to the middle of the incisor overlap. Steiner relates it to S-N (about 14°), Tweed to Frankfort (about 10°, from 8 to 12°). Never mix the two.
• N-A and N-B: they locate each jaw and serve as references for the incisors (U1-NA, L1-NB).
• A-Pog: the line from point A to Pogonion, used to assess the convexity of the skeletal profile and, in Ricketts’ analysis, the position of the incisors.
• Palatal plane (ANS-PNS) and the incisor long axes (from incisal edge to apex).
• Esthetic lines: Steiner’s S line (from soft-tissue pogonion to the columella of the nose) and Merrifield’s Z line (from the most prominent lip to soft-tissue pogonion).

Measurements and norms

These are the norms used by our free tool. They follow the Steiner and Tweed norms of university course handouts; standard deviations may vary slightly from one program to another.

Three habits make the reading more reliable:

• check the Tweed triangle: FMA, IMPA and FMIA add up to 180°. If the total is wrong, one measurement is wrong;
• read measurements in pairs: ANB is confirmed by the Wits, FMA by SN-GoGn, and an incisor angle by the corresponding distance;
• speak of dentoalveolar protrusion only when the angle and the distance of the incisor are both increased (retrusion when both are decreased). An increased angle alone describes labial tipping.

For the lower incisor, Tweed’s rule turns the divergence into a target: with an FMA of about 22° to 28°, aim for an FMIA of 67°; above 28°, aim for 65°, the incisor being more upright in the hyperdivergent patient; below 22°, do not exceed an IMPA of about 92°. The cut-offs vary between courses: use the version taught in your program. The Steiner side is detailed in our guide to the Steiner analysis, and the Tweed triangle in our Tweed article (in French).

These norms were established on specific samples, largely North American. A value "outside the norm" is a signal to explain, not an indication for treatment.

The most common tracing errors

• Measuring before looking: an abnormal sella turcica or vertebrae go unnoticed if you skip the qualitative reading.
• Uncalibrated millimeter measurements: a tracing left in pixels makes U1-NA and L1-NB in millimeters, the Wits and the facial heights impossible to compare with the norms. Only angles and ratios remain readable.
• Mixing occlusal planes: a value measured on S-N carried into an analysis that expects it on Frankfort skews everything that follows.
• Not checking the Tweed triangle: a sum other than 180° often reveals two angles swapped.
• Mishandling a double mandibular contour: if the two borders do not cross, the usual approach is to trace the midline between them, and to note a possible sign of asymmetry.
• A misplaced incisor apex: an error of a few millimeters at the apex changes the long axis of the tooth by several degrees.

Interpreting without being misled

ANB is convenient, but it depends on the position of Nasion, the inclination of S-N and the vertical dimension. A rotation of the mandible can change ANB without any change in the relationship between the jaws. This is why it is confirmed with the Wits appraisal, which is little influenced by the vertical dimension. A classic teaching example: an ANB of 4° (upper limit) with a Wits of −1 mm remains a borderline Class I, not a Class II. The Wits has its own weakness: it depends on how the occlusal plane is traced, which is tricky in the mixed dentition.

The teeth often compensate for the skeletal discrepancy. In a skeletal Class II, the lower incisor is frequently proclined and the upper incisor upright; in a Class III, the reverse. These compensations hide the size of the discrepancy in the mouth, and they shape the strategy: keep them for camouflage, or remove them before surgery.

Finally, the tracing must explain what you see chairside. A Class II ANB in a patient with a harmonious profile deserves a second look at the landmarks.

AI-assisted tracing: useful, but check every landmark

Software can now place the landmarks automatically in a few seconds. The gain in time and reproducibility is real, but the algorithm tends to go wrong where the human eye also hesitates: Porion, Orbitale, Gonion, incisor apices, points A and B. Every landmark placed by AI must be checked before the measurements are used for a diagnosis. Our article on artificial intelligence in cephalometrics (in French) details how it works and where it fails.

Practice tracing for free online

Cephalometrics is learned by tracing. Our online cephalometric analysis is free, with no sign-up: you load a lateral cephalogram, calibrate it, place the 15 landmarks, and the Steiner and Tweed measurements are calculated with their interpretation. The image stays on your computer. Then you can carry ANB, U1-NA and L1-NB over to the Steiner’s chevrons calculator, and complete the space analysis on the arches with the space analysis and Bolton analysis tools.

A good exercise: deliberately move Nasion or Porion by a few millimeters and watch the effect on the measurements.

Frequently asked questions

Which cephalometric analysis should I choose?

There is no single analysis. Steiner focuses on the sagittal discrepancy and the incisor position; Tweed on the divergence and the lower incisor. The two complement each other. What matters is knowing the references of the analysis you use, and not mixing its norms with those of another.

What is the normal ANB value?

The normal ANB is 2° ± 2°. In Ballard’s classification, an ANB between 0 and 4° corresponds to a skeletal Class I, above 4° to a Class II and below 0° to a Class III. ANB should always be confirmed by the Wits appraisal and by examining the profile.

Do I need to calibrate the radiograph?

Only for distances. Angles do not depend on the scale, but U1-NA and L1-NB in millimeters, the Wits and the facial heights do. Calibration means measuring the ruler visible on the cephalogram: click both ends of a known distance and enter its length.

Can I trace a cephalogram online for free?

Yes. Our online cephalometric analysis tool is free, with no sign-up or installation. It calculates SNA, SNB, ANB, SN-GoGn, the incisor position and the Tweed triangle from the landmarks you place, and the radiograph is not sent anywhere. It is a tool to learn and to check a tracing; in practice, complete software saves the tracing in the patient’s record and links the measurements to the diagnosis.

Can I save this guide as a PDF?

Yes: use your browser’s "Print" function, then "Save as PDF". For the official norms of your program, refer to your university’s course handout: some values vary from one program to another.

With GS Ortho

In GS Ortho, AI places 29 landmarks on the lateral cephalogram, soft-tissue profile included, and the Steiner and Tweed analyses are calculated straight away. You check each landmark, correct it if needed, then validate: the decision stays yours. The tracing goes into the patient’s record, the measurements feed into the diagnostic studio, and an orthodontist’s opinion can be requested in one click, straight from the case. The GS Ortho interface is currently available in French.

Book a demo

cephalometric tracing lateral cephalometric radiograph cephalometric landmarks steiner tweed norms