Image Fusion

Image Fusion

Image fusion overlays a functional series — PET (PT) or SPECT (NM) — on top of an anatomical CT or MR base, so metabolic uptake can be read against the underlying anatomy. The overlay is a registration-free geometric fusion: Weasis aligns the two series from their DICOM spatial metadata alone, with no manual or algorithmic registration step. Available since Version4.7.1.

Note

Fusion is a pure spatial overlay driven by each image’s position in the patient coordinate system. It does not deform or re-register the images, so it is only offered when the two series are genuinely co-located (see Requirements below).

Requirements

A series can be fused onto the displayed base only when all of the following hold:

Condition Detail
Modality pairing A functional overlay (PT, NM) on an anatomical base (CT, MR).
Same study Both series share the same StudyInstanceUID.
Volume geometry Both are cross-sectional volumes (≥ 2 slices with Image Position / Orientation Patient and Pixel Spacing).
Same coordinate system Either the FrameOfReferenceUID matches, or — as is common for separately reconstructed PET/CT — the two volumes overlap in patient space.

When no compatible overlay is found in the study, the fusion controls stay disabled.

Enable fusion

The fusion controls live in the Fusion section of the Image tool panel. The section is collapsed by default — click its header to expand it.

  • Enable Fusion — turns the overlay on or off. The other controls become active only while fusion is enabled.
  • Series — the functional series to overlay. The list contains every compatible overlay found in the current study (see Requirements).
  • LUT — the color lookup table applied to the overlay. PET (a hot-metal palette tuned for functional data) is selected by default. The overlay is colorized through the functional series’ own window/level, so it shows exactly the colors the overlay would have if displayed directly with that LUT.
  • Opacity — two sliders set a flat blend (cross-fade) at composite time: result = baseOpacity·base + overlayOpacity·overlay. Each is labelled with the modality of the layer it controls (e.g. CT for the base, PT for the overlay). Defaults are 100 % base and 75 % overlay. Setting base 0 % and overlay 100 % shows the pure colorized overlay, identical to the functional series viewed on its own with the same LUT.

The overlay follows the base view as you scroll, zoom, window, or reslice it. Because the fusion is resampled from the functional volume, it stays correct on oblique, coronal and sagittal planes, not only the native acquisition plane.

Note

The Reset action turns fusion off and returns to the plain base image, like any other display setting it restores.

SUV statistics on a region of interest

When the overlay is a PET series carrying the required metadata, a closed measurement drawn on the fused base image also reports the overlay’s SUV values inside the region — Min, Max and Mean, in SUVbw, g/ml. These rows are sampled from the original PET voxels (no resampling loss on the maximum) and are tagged with the overlay’s modality — for example Max (PT) — so they read alongside the base-image statistics.

Tip

SUV is computed with the body-weight method (SUVbw), following the vendor-neutral QIBA definition. The same values can be read directly on the PET series itself — see SUV measurements.

Fusion in the MPR viewer

Fusion on MPR Fusion on MPR

Opening the MPR viewer from a fused 2D view carries the fusion over: the three MPR planes start with the same overlay series, LUT and opacities you set in the 2D view. Each plane keeps its own fusion settings, so you can fine-tune — or disable — the overlay independently per plane afterwards, without affecting the 2D view or the other planes.

Note

Inheritance only applies when the overlay is still compatible with the reconstructed MPR base. If you open MPR from a non-fused view, the planes start with fusion off, and you can enable it from the same Fusion section as in the 2D viewer.

See also