# Airway volume on CBCT and the narrowest cross-section

> A CBCT airway volume depends on where the airway starts and stops and how it is outlined, so two programs can give different numbers for one scan.

- URL: https://jawviewer.com/blog/airway-volume-cbct-what-it-means
- Category: Airway and sinuses
- Published: 2026-11-21
- Updated: 2026-11-21
- Author: Jaw Viewer
- Reading time: 5 min
- Tags: cbct, airway, measurements

## Key points

- In a 2010 study of 30 CBCT scans, three DICOM viewers repeated airway volumes reliably, but accuracy against manual outlining was poor, which points to systematic errors.
- In a 2025 lab test, four programs measured one 3D-printed airway model ten times, and their accuracy differed significantly (P < .001).
- A 2011 review of 46 articles found 3D airway analysis can be accurate and reliable, with open problems in breathing phase, tongue position, jaw shape and where the airway ends.
- The narrowest cross-section is the smallest slice through the outlined airway, so it moves whenever the outline or its ends move.

An airway volume on a CBCT scan is the amount of air inside a region that a program has outlined in the throat, in cubic centimetres. The number depends on how the airway was outlined and where it was cut off at the top and bottom, so it describes that outline more than it describes the person. The volume and narrowest cross-section both depend on those choices, which helps explain why programs disagree on the same scan.

## What is outlined?

The airway is the open channel behind the nose and mouth that runs down to the voice box. On a scan it shows up as dark air between the grey soft tissue. Outlining it is called segmentation: the program (and often the user) marks which voxels belong to the airway and leaves the rest out.

The outline needs ends. The nose side joins the nasal cavity and the sinuses, and the lower end continues past the tongue towards the windpipe. Someone has to choose where the measured airway starts and stops. A 2011 systematic review of 46 articles lists the definition of the anatomical boundaries of the upper airway among the obstacles still to be solved, along with the breathing phase and the position of the tongue and jaw.[1] The same review concluded that 3D airway analysis on CBCT can be done accurately and reliably, so the method is sound. The open questions are about what exactly is being measured.

*Figure: The airway overlay in the slices of the public sample scan, next to its volume, its narrowest cross-section and a chart of its width. In this sample scan the overlay shows what is counted. The chart runs along the airway. Sample scan "CBCT-MR Head (DZ-CBCT)", 3D Slicer sample data, donated for use without restrictions. Screenshot from Jaw Viewer.*

## How are the volume and the narrowest area worked out?

Both numbers come from the outlined voxels. A voxel has a known size in millimetres, so the volume is the number of airway voxels multiplied by the volume of one voxel. With 0.3 mm voxels, one voxel is 0.027 mm³, and a volume of 1 cm³ (1000 mm³) is about 37,000 voxels (our arithmetic). Because the count is large, the edge matters most: a voxel half filled with air and half with tissue has to be counted in or out, and that choice repeats along the whole surface. The voxel size itself is covered in [CBCT voxel size and sharpness](https://jawviewer.com/blog/cbct-voxel-size-sharpness.md).

The narrowest cross-section starts from a line along the airway. The program cuts the outlined airway into slices at right angles to that line and measures the cross-sectional area of each one, in square millimetres. The smallest value is the narrowest area, and a chart of width or area along the airway shows where it sits.

Moving the top or bottom border changes the volume because there is more or less airway to count. Move the outline and the narrowest area can change too, since it is a minimum over a stretch whose ends and edges you chose.

## Why do methods give different numbers?

Programs differ in how they decide that a voxel is air, and in how much the user steers the result. Studies have tested how much this affects the result.

In a 2010 study, 30 CBCT scans were measured in three commercial DICOM viewers and compared with a manual segmentation program. The measurements were repeated after two weeks, and reliability was high for all programs, and results correlated highly with the manual program. Even so, the volumes differed significantly from the manual program, and the programs were not consistent with each other. The authors read this as high reliability with poor accuracy, which suggests systematic errors.[2]

A 2025 laboratory test removed the person from the question. One airway model was 3D printed from CBCT data and scanned 10 times under identical conditions. Four programs measured it by semi-automatic segmentation, and the results were compared with the volume of the model's cavity filled with water. The accuracy of the volume differed significantly between programs (P < .001), and two of the four came closest to the water measurement.[3] That is one model in one lab, so it shows that programs can disagree. It does not rank them for a living airway.

> Repeatable is not the same as right
>
> A program can return nearly the same number every time and still be off by a steady amount. Comparing two scans with the same program and the same borders removes some of that error. Comparing a number from one program with a number from another does not.

## Why does one number say little about one person?

A single volume or narrowest area has no context. It does not say how the borders were chosen, whether the person was breathing in or out, or where the tongue and jaw were during the scan. The review lists respiration phase, tongue position and mandible morphology as factors that still affect upper airway evaluation on CBCT.[1] A scan takes a moment, so the airway it shows is the airway at that moment.

How far values spread between people, and what that means for a "typical" number, is covered in [how much upper airway volume varies between people](https://jawviewer.com/blog/upper-airway-volume-normal-values-variation.md). An airway volume describes the outlined region.

## How to read the result in Jaw Viewer

In Jaw Viewer, the Airway tool outlines the airway in the slices and in 3D and reports its volume and its narrowest cross-section, with a chart along the airway. The [airway and sinuses guide](https://jawviewer.com/guide/airway-and-sinuses.md) walks through the steps. A few habits make the number easier to interpret.

- Look at the overlay in every slice direction before you read a value, and check the top, bottom and sides of the region.
- If the outline includes the wrong space or leaves out part of the airway, adjust the region controls and read the numbers again.
- Note how the region was chosen. A difference between two numbers can come from the selection as much as from the scan.
- Treat millimetre readings with the same caution as any other measurement on a scan, as described in [how accurate CBCT measurements are](https://jawviewer.com/blog/how-accurate-are-cbct-measurements.md).

The sinuses are outlined the same way and have the same boundary question; see [maxillary sinus volume and the range reported in studies](https://jawviewer.com/blog/maxillary-sinus-volume-normal-range.md). For what a scan can and cannot show in general, start with [what a CBCT scan shows](https://jawviewer.com/blog/what-a-cbct-scan-shows.md).

This is general information, not medical advice. Jaw Viewer is not a medical device. Talk to your surgeon, orthodontist or radiologist about your own case.

## Questions and answers

### Why do two programs give different airway volumes for the same scan?

They outline the air differently and may place the top and bottom of the airway differently. In a 2025 lab test, four programs differed significantly in accuracy on one printed airway model.

### What is the narrowest cross-section of the airway?

It is the smallest area among the flat slices taken along the outlined airway, from top to bottom. It depends on the outline and on where the airway is cut off at each end.

### Does the airway change shape between scans?

It can. A 2011 review lists breathing phase, tongue position and jaw shape among the open problems in airway analysis on CBCT, so one scan shows one moment.

## Terms used

- **Segmentation**: Outlining a structure in a scan, here the air inside the throat, so that its volume and shape can be measured.
- **Voxel**: A three-dimensional pixel. A scan is a grid of voxels, each with one grey value.
- **Cross-sectional area**: The area of a flat slice through a structure, at right angles to its length. For the airway it is the open space at one height.

## Sources

1. Guijarro-Martínez R, Swennen GR (2011). Cone-beam computerized tomography imaging and analysis of the upper airway: a systematic review of the literature. Int J Oral Maxillofac Surg, vol. 40, pp. 1227-37. https://doi.org/10.1016/j.ijom.2011.06.017 46 articles included out of 382 screened.
2. El H, Palomo JM (2010). Measuring the airway in 3 dimensions: a reliability and accuracy study. Am J Orthod Dentofacial Orthop, vol. 137(4 Suppl), pp. S50.e1-9. https://doi.org/10.1016/j.ajodo.2010.01.014 30 CBCT scans, three DICOM viewers compared with a manual segmentation program.
3. Naralan ME, Köse TE, Gonca M et al. (2025). Accuracy of upper airway volume measurements using different software products: a comparative analysis. Dentomaxillofac Radiol, vol. 54, pp. 350-356. https://doi.org/10.1093/dmfr/twaf023 One 3D-printed airway model scanned 10 times; a laboratory result, not a patient study.

---

Jaw Viewer (https://jawviewer.com). Page index for AI tools: https://jawviewer.com/llms.txt. Not a medical device.
