Learn to read CBCT

Anatomy

Bifid mandibular canal and other canal variants on CBCT

After this page you will be able to recognise a bifid mandibular canal, a retromolar canal and an accessory mental foramen on serial cross-sections, and explain why none of them can be confirmed from a single slice.

The usual pattern first

The common course of the inferior alveolar canal is one tunnel from the mandibular foramen on the inner surface of the ramus, through the body below the molar roots, to the mental foramen on the outer surface below the premolars. If that is not yet familiar, read finding the mandibular canal first, because every variant here is recognised by comparing it with that pattern. Variants are common enough to look for on every scan, and CBCT finds many that a panoramic radiograph blurs into the main canal.

Bifid canals

A bifid canal is one that splits into two channels for part of its length. The second channel may run parallel to the main canal, above or below it, and rejoin after a few centimetres; branch up and forward towards the molar roots; or branch backwards and upwards towards the retromolar region, the retromolar canal described below. A canal with three channels is called trifid and is rare.

In cross-section each channel looks like the main canal, a dark dot with a thin bright rim, so a bifid canal shows two dots instead of one, the second usually smaller. On a sagittal slice or the panoramic image the two show as two dark bands with corticated outlines, one above the other or diverging.

The retromolar canal and accessory mental foramina

The retromolar canal leaves the mandibular canal near the ramus or the angle and runs up and forward to open at the retromolar foramen, a small opening on the bone behind the last molar, in the retromolar triangle. On axial slices at the level of the molar crowns the foramen is a small gap in the cortex behind the third molar; scroll down and the channel runs backwards and down to join the main canal. Marrow spaces behind a third molar, and the follicle of an impacted one, can look like a short channel on one slice (see impacted third molars on CBCT), so apply the test for every variant: does it persist, does it have a rim, and does it connect to the main canal?

Some people have one or more small extra openings near the mental foramen, usually within a centimetre of it, each with its own short canal joining the main canal. On an axial slice an accessory foramen is a second, smaller gap in the buccal cortex near the main one. Distinguish it from a nutrient canal, which passes through the cortex but does not connect to the mandibular canal: scroll back from the opening, and an accessory canal leads to the main canal while a nutrient canal fades into the marrow.

Recognising variants on serial cross-sections

The reliable way to find any of these is a series of cross-sections perpendicular to the arch, stepped a few millimetres at a time, because such a slice cuts the canal and any branch almost at right angles. In JawAtlas that is the Pano stage. On a fresh install the first visit to the Pano stage opens the tracing view: tap at least four points along the arch on the axial slice and tap "Show panoramic" (see how to trace the dental arch). Then drag the yellow scout line along the panoramic image and watch the cross-section below it, with its readout of millimetres along the arch.

Start at the ramus and move forward slowly. The main canal is one dark dot with a rim. When a second dot appears, step backwards and forwards: a real branch is there on several consecutive cross-sections, keeps a rim, and somewhere leaves the main canal as a figure of eight. Follow it until it rejoins, reaches a root or opens on a surface, saving a moment where it leaves, where it is clearest and where it ends (see saving and replaying moments).

Then confirm in the MPR stage: put the white crosshair on the second channel in a coronal slice and switch to axial and sagittal. The same dark channel should be there in both. A branch that exists in one plane only is not a branch (see how the three planes link).

Why single slices mislead

A canal is a tube and a slice is a plane, so what you see is the cut between them. Where the tube curves, a flat slice can cross it twice, and one canal appears as two dark spots. Where the slice runs along a tube obliquely, a short stretch becomes a long smear that looks like a second channel beside the first. A large marrow space with a partly corticated wall is a dark spot with a rim, indistinguishable from a canal on one slice. And where the rim of a real branch is thin, the voxel grid averages it away, so a genuine channel can vanish on one slice and reappear on the next (see field of view and voxel size).

The panoramic image adds a trap of its own. It is built by averaging a slab along the arch, so a branch that lies buccal or lingual of the slab is lost, and two structures at different depths can be averaged into one band (see how the panoramic view is built). Use the panoramic image for the overview and the cross-sections to decide.

Things that look like variants and are not

The mylohyoid groove runs down and forward from the mandibular foramen on the inner surface of the ramus. On an axial slice it is a notch in the lingual cortex, open to the soft tissue, not a tunnel inside the bone. The mandibular incisive canal is the faint continuation of the main canal forward of the mental foramen, part of the normal pattern rather than a bifid canal. The lingual foramen is a small midline opening on the inner surface below the incisors, with its own short canal.

JawAtlas is an education and training aid. It is not a medical device and must not be used for diagnosis, treatment planning or any clinical decision.

Last reviewed 2026-10-14