9.1 3D Rotational Angiography & Cone-Beam CT (CBCT)

Key Takeaways

  • Model presets determine the arc, duration and sampling. A short-scan fan-beam rule does not establish exact circular cone-beam reconstruction.

  • FDK provides approximate reconstruction. Isotropic voxel spacing does not guarantee isotropic resolving power.

  • A matched mask and fill can suppress background for 3D vascular review. Compare with other views when detail is uncertain.

  • Post-thrombectomy on-table CBCT helps assess the brain but may miss lesions or leave iodine-versus-blood uncertainty.

  • Bullseye alignment and progression imaging answer different spatial questions. Verify registration and actual needle tip location.

Last updated: November 2026

9.1 3D Rotational Angiography & Cone-Beam CT (CBCT)

Clinical Core: A rotating C-arm acquires multiple projected views of the same region. Reconstruction can reveal vascular or tissue relationships that are difficult to see in one projection. The result depends on the selected program, patient position, filling, registration and the reconstruction method.

Acquisition geometry

Plan the acquisition around the target and clinical question. Center the relevant anatomy, select a field that includes it across the sweep, and confirm clearance of the patient, supports, shields, cables and other gantry. Use the manufacturer's clearance checks for that program. Confirm the programmed angle and detector motion before exposure. Patient movement can cause blur or misregistration. Isocenter positioning alone cannot ensure that a broad target stays inside every projection.

The X-ray tube provides exposures at successive angles and the area detector records projections. For an illustrative constant sampling rate, 60 frames per second for 5 seconds gives 300 sampled frames. The Philips Azurion Release 3.1 instructions describe a low-dose head CBCT program at 60 frames per second over 5 seconds. Do not transfer that selection or extrapolate its dose to another scan, system or patient. Frame count is one factor, along with the mode and dose per exposure.

For the two-dimensional fan-beam short-scan geometry, a data interval of 180° plus the full fan angle can cover the necessary rays in the central plane. In that example, a 30° full fan angle gives 210°. Cone-beam reconstruction from a circular orbit can still have incomplete information away from the central plane. This formula does not establish an exact universal 3D reconstruction or an equipment travel requirement.

Reconstruct and display the volume

The Feldkamp–Davis–Kress (FDK) algorithm is an approximate filtered backprojection approach for circular cone-beam geometry. It geometrically weights the measurements, filters detector rows and backprojects the results through the volume. Other processing and reconstruction approaches may be used. A smaller reconstructed voxel or an isotropic voxel grid permits flexible reformatting. Neither guarantees that true resolution is equal in every direction or to the voxel size.

DisplayWhat it showsLimitation
Multiplanar reformation (MPR)A slice through the volume, including an oblique plane through a structure.Confirm slice orientation and measurement scale.
Maximum intensity projection (MIP)Maximum voxel value along each viewing ray.A bright structure can obscure another structure or conceal depth.
Volume rendering (VR)A projected view with assigned opacity and color to illustrate the volume.Display settings can hide a small branch or alter apparent borders.

Match the acquisition to the question

  • Native and subtracted rotational imaging. A native rotational acquisition can show vessel and tissue context with or without contrast, according to the program. For a dual-pass subtracted study, record a mask and a fill sweep at matched angular positions. The system can align and subtract projection data to reduce background before reconstruction. Poor filling, motion between passes, scatter, truncation and metal can reduce reliability. Set injection and timing to opacify the target for the needed views under the applicable catheter and equipment limits.
  • Aneurysm planning. Multiple views and reformations of a reconstructed cerebral angiographic volume can show neck and branch relationships and help choose a working projection. Compare with native or 2D angiographic views to resolve a questionable small branch. Rotating a rendered model cannot recreate structures missing from the collected data.
  • On-table assessment after thrombectomy. Flat-panel CBCT can provide early cross-sectional information without moving the patient. Compared with follow-up CT, primary studies show imperfect detection of intracranial hemorrhage and ischemia. After recent iodine administration, hyperattenuation may represent iodine, hemorrhage or both. An uncertain or reassuring appearance must be interpreted alongside clinical findings and the team's follow-up imaging plan.
  • Hepatic embolization. A contrast-enhanced acquisition can assess feeders, tumor enhancement and perfused territories. Selected dual-phase studies use early information about arterial supply and later information about tissue enhancement. Timing depends on the clinical target and contrast delivery. An extrahepatic enhancement pattern can prompt reassessment before treatment. Vascular enhancement does not measure actual radioactive microsphere distribution or absorbed dose. Radioembolization has separate mapping and verification steps.

Virtual needle guidance

A guidance workstation can mark a skin entry and internal target on a registered planning volume. Verify that the intended path clears adjacent structures and that the patient's position and respiration still match the plan. The overlay is a guide to the planned path. Tissue motion, needle deflection and registration error remain possible.

A bullseye view looks along the planned path, so the projected entry and target can coincide. It helps judge directional alignment. A different, non-collinear progression view shows the shaft and progress of the tip along the planned path. It need not be exactly 90°. For example, a needle may project over the target in a bullseye view while its tip remains short of the target. Inspect depth from another view and verify actual position with appropriate imaging when the procedure requires it.

Loading diagram...
3D Rotational Angiography, CBCT & Virtual Needle Navigation Architecture
Test Your Knowledge

What does a reconstructed rotational angiographic study add when standard projections overlap the neck of an intracranial aneurysm?

A

It guarantees that contrast, motion and display choices cannot affect the result.

B

It permits review at additional virtual angles and can clarify neck and branch relationships when paired with the source views.

C

It directly measures blood velocity everywhere in the aneurysm.

D

It replaces the need for angiographic contrast with magnetic resonance.

Test Your Knowledge

Immediately after thrombectomy, a flat-panel CBCT shows a focal hyperattenuating region in a patient who received iodine. What can the team conclude from this finding alone?

A

It proves hemorrhage without need for clinical review.

B

It proves benign contrast staining because the patient received iodine.

C

It flags a finding for evaluation. Iodine, hemorrhage or both can contribute, so correlate with the clinical findings and follow-up imaging plan.

D

It provides a direct measurement of cerebral blood flow velocity.

Test Your Knowledge

Entry and target markers overlap in the bullseye view of a registered needle plan. Which statement is correct?

A

This view supports trajectory alignment. Check depth and actual tip position using an appropriate second view or cross-sectional imaging.

B

Overlap proves that the tip has reached the target.

C

Overlap substitutes for checking patient movement after planning.

D

The system always requires exactly 90° of gantry motion before any further image.

Sections you finish are checked off in the contents.