Cephalometric analysis is a standardized method for measuring craniofacial skeletal and dental relationships from a lateral cephalogram — a specific type of radiograph taken from the side of the head with the patient in a fixed orientation. Orthodontists use cephalometric analysis to assess skeletal pattern (Class I, Class II, Class III), facial proportions, and dental positions before, during, and after orthodontic treatment. It is one of the oldest standardized diagnostic methods in orthodontics, dating back to the 1930s, and remains central to modern orthodontic case planning.
The analysis works by identifying anatomical landmarks on the cephalogram — specific points like Sella, Nasion, A point, B point, Pogonion, and Menton — and computing measurements between them: angles (SNA, SNB, ANB, Frankfort-Mandibular Plane Angle) and linear distances. These measurements compare to established norms and to the patient's own measurements at other time points. The pattern of measurements informs the orthodontist's treatment plan, the choice of mechanical approach, and the expected outcomes.
Orthodontic treatment is mechanical work on teeth and jaws over months or years. Without cephalometric measurements, the orthodontist is working from clinical impression alone — looking at the patient's profile and intraoral situation and inferring the underlying skeletal pattern. Cephalometric measurements turn that inference into structured data: the patient has an ANB of 6° (indicating a Class II skeletal pattern); the maxilla is protrusive; the mandible is retrusive; the lower incisors are proclined relative to the mandibular plane. The measurements drive the case plan in a way the clinical eye alone cannot.
Cephalometric analysis also drives outcome assessment. When the orthodontist completes treatment, a post-treatment cephalogram measures the same landmarks against the pre-treatment values. The changes are quantifiable: ANB reduced from 6° to 3°; lower incisors retracted by 4mm; facial profile improved by measurable angle. Without cephalometric measurements, outcome claims rest on clinical impression and photographic comparison; with them, outcomes are documented in standardized form across time.
The third reason cephalometric matters is comparison across the patient's own treatment arc. An orthodontic case that runs for 24 months may have cephalograms at T0 (start), T1 (mid-treatment), and T2 (completion), with each set of measurements compared against the others to track treatment progress. This longitudinal view is essential for cases that are not progressing as expected and need mid-course adjustment.
The standardized vocabulary of cephalometric work.
Sella (S, the midpoint of the sella turcica), Nasion (N, the most anterior point of the frontonasal suture), A point (the deepest point on the maxillary alveolar bone), B point (the deepest point on the mandibular alveolar bone), Pogonion (Pog, the most anterior point of the chin), Menton (Me, the lowest point of the mandibular symphysis), Gonion (Go, the most posterior point of the angle of the mandible), Basion (Ba, the most anterior point on the foramen magnum). These define skeletal relationships.
Upper incisor edge and apex, lower incisor edge and apex. These define dental relationships to the skeletal base — proclination, retroclination, vertical positioning. Dental landmarks drive much of the orthodontic case-planning decision about how to move teeth relative to the underlying skeletal frame.
SNA (the angle from Sella to Nasion to A point — indicating maxillary position), SNB (Sella-Nasion-B point — indicating mandibular position), ANB (the difference, indicating skeletal Class I/II/III pattern). FMA (Frankfort-Mandibular Plane Angle) describes vertical facial pattern. These angles are the most commonly referenced cephalometric measurements in clinical conversation.
Developed in the 1950s by Cecil Steiner. Uses SNA, SNB, ANB, and dental measurements relative to the SN plane. One of the most widely-used methods in U.S. orthodontic training. Steiner emphasizes the relationship of teeth and skeletal bases to the cranial base.
Developed by Charles Tweed. Uses the Frankfort horizontal plane as a reference. Tweed analysis emphasizes the position of the lower incisors relative to the mandibular plane and Frankfort horizontal — measurements that drive specific extraction and non-extraction treatment decisions.
The Downs analysis (developed at Northwestern in the 1940s) uses ten measurements covering both skeletal and dental relationships. The Vertical analysis focuses on vertical facial dimensions. The Eastman analysis is widely used in U.K. orthodontic training. Each method emphasizes different aspects of the same underlying anatomy; many orthodontists use different methods for different case types.
Modern cephalometric AI handles the landmark identification step — historically the most time-consuming part of the workflow — within seconds. The AI identifies anatomical landmarks on the uploaded cephalogram with a per-landmark confidence score, and the orthodontist reviews and validates the detections before measurements are finalized. The clinical decision-making — interpretation of the measurements, treatment planning, mechanical approach — remains the orthodontist's.
WIO CLINIC's cephalometric AI supports all six standard analysis methods (Basic, Steiner, Tweed, Downs, Vertical, Eastman) with method selection at the case level. Per-landmark confidence scores are visible to the practitioner. The validation workflow is explicit. AI is positioned as clinical decision support — every AI output is reviewed and validated by a clinician before clinical action.
Orthodontists routinely perform cephalometric analysis on every case. Some general dentists trained in orthodontic case management also use it. Oral and maxillofacial surgeons reference cephalometric analysis for surgical planning in orthognathic cases.
The exact number varies by analysis method. Steiner uses around a dozen primary landmarks; comprehensive analyses like Downs reference more. Modern AI-assisted analysis typically identifies a comprehensive set so that any of the standard methods can be computed from the same landmark identification.
Cephalometric analysis produces measurements that inform clinical decisions. It is not in itself a diagnostic test in the sense that it does not autonomously produce a diagnosis. The orthodontist interprets the measurements in the context of the patient's clinical examination, photographs, and other diagnostic records to arrive at a treatment plan.
Manual analysis: the orthodontist identifies landmarks on the cephalogram by hand, typically a 15-30 minute process per case. AI-assisted analysis: the AI identifies landmarks within seconds, with per-landmark confidence scores; the orthodontist reviews and validates the detections before measurements are finalized. Both produce the same kind of output; AI compresses the time required while preserving the orthodontist's review.