Accuracy of 3D volumetric image registration based on CT, MR and PET/CT phantom experiments

Abstract

Registration is critical for image-based treatment planning and image-guided treatment delivery. Although automatic registration is available, manual, visual-based image fusion using three orthogonal planar views (3P) is always employed clinically to verify and adjust an automatic registration result. However, the 3P fusion can be time consuming, observer dependent, as well as prone to errors, owing to the incomplete 3-dimensional (3D) volumetric image representations. It is also limited to single-pixel precision (the screen resolution). The 3D volumetric image registration (3DVIR) technique was developed to overcome these shortcomings. This technique introduces a 4th dimension in the registration criteria beyond the image volume, offering both visual and quantitative correlation of corresponding anatomic landmarks within the two registration images, facilitating a volumetric image alignment, and minimizing potential registration errors. The 3DVIR combines image classification in real-time to select and visualize a reliable anatomic landmark, rather than using all voxels for alignment. To determine the detection limit of the visual and quantitative 3DVIR criteria, slightly misaligned images were simulated and presented to eight clinical personnel for interpretation. Both of the criteria produce a detection limit of 0.1 mm and 0.1 degree. To determine the accuracy of the 3DVIR method, three imaging modalities (CT, MR and PET/CT) were used to acquire multiple phantom images with known spatial shifts. Lateral shifts were applied to these phantoms with displacement intervals of 5.0+/-0.1 mm. The accuracy of the 3DVIR technique was determined by comparing the image shifts determined through registration to the physical shifts made experimentally. The registration accuracy, together with precision, was found to be: 0.02+/-0.09 mm for CT/CT images, 0.03+/-0.07 mm for MR/MR images, and 0.03+/-0.35 mm for PET/CT images. This accuracy is consistent with the detection limit, suggesting an absence of detectable systematic error. This 3DVIR technique provides a superior alternative to the 3P fusion method for clinical applications.

Figure 1

The flow chart of 3D volumetric image registration process. The 4 volumetric image data are stored in a 32‐bit voxel buffer array, which can be retrieved and manipulated in the two different processes: registration transformation and volumetric visualization. The registration process iterates until the registration criterion is satisfied visually and/or quantitatively.

Figure 2

Demonstration of ray‐casting algorithm for volumetric visualization. (a) The relationship of voxels in an image volume and pixels in an image plane and (b) RGBA accumulation along a ray until the accumulated opacity reach unity.

Figure 3

Volumetric views of two identical CT phantom images with simulated spatial shifts. Top row (A to C): translational shifts (Xt) of 0.5, 0.2 and 0.0 voxels ($\text{voxel}=0.78\text{mm}$) were applied to the aligned image (C); Bottom row (D to F): rotational shifts (Xr) of 0.5°, 0.2° and 0.0° were applied to the aligned image (F). The color homogeneity on the “skin” landmark improves as the alignment is improved. The translational (lateral) shifts appear mostly on the left and right sides of the image volume: the larger the surface grayscale gradient and the larger the surface oblique angle (between the ray and surface normal), the larger the visual color inhomogeneity would be. The rotational (around the superior‐inferior axis through the center of the image volume) shift causes non‐uniform displacements in the directions perpendicular to the rotational axis: the larger the distance of the viewing voxel to the rotational axis, the bigger the rotational displacement and so the more dramatic color inhomogeneity.

Figure 4

Three head phantoms for (A) CT, (B) MR and (C) PET/CT experiments. Head holders and tapes were used to immobilize the phantoms, and graph papers and magnifying glass were used for phantom positioning. The finest line width of 0.13 mm was used to align two gridlines on the couch and the phantom holder. The alignment was checked on all four sides of the phantom holder for a translational shift.

Figure 5

Quantitative criteria vs. spatial shifts (translation or rotation). (A) Identical CT/CT image alignment: VAR criterion is used in lateral direction (Xt, translation) or lateral axis (Xr, rotation). Legends: (a) anterior and (b) superior views of Xt translational shifts, and (c) anterior and (d) superior views of Xr rotational shifts. (B) Co‐registered PET/CT image alignment: mVAR criterion is used in lateral direction (Xt, translation). From (a) to (h), eight curves of the 4 set PET/CT images based on superior views (solid symbols) and anterior views (open symbols) are shown.

Figure 6

Anterior views of PET/CT images with lateral off‐alignments: (A) $-0.5$ voxels (mm), (B) 0.0 voxel (mm) and (C) $+0.5\text{mm}$ voxels (mm). The arrows point to the region with color inhomogeneity. Note: the local color inhomogeneity shown in (B) is caused by different imaging resolutions in PET and CT and slightly different R/G‐LUTs on the two image histograms.

Figure 7

Using bony anatomy as registration landmark. The original two CT images (A, C) are $15.0\pm 0.1\text{mm}$ apart laterally and the registered images (B, D) are obtained with a lateral shift of 15.00 mm. A minor superior‐inferior shift of 0.3 voxel (0.23 mm) is made to compensate the couch positioning error. Note: the local color inhomogeneity (contour pattern) shown in (B, D) is caused by limited imaging resolution and visualization with slightly different R/G‐LUT settings, rather than global image misalignment.