EM reconstruction of dual isotope PET using staggered injections and prompt gamma positron emitters

Abstract

Purpose: The aim of dual isotope positron emission tomography (DIPET) is to create two separate images of two coinjected PET radiotracers. DIPET shortens the duration of the study, reduces patient discomfort, and produces perfectly coregistered images compared to the case when two radiotracers would be imaged independently (sequential PET studies). Reconstruction of data from such simultaneous acquisition of two PET radiotracers is difficult because positron decay of any isotope creates only 511 keV photons; therefore, the isotopes cannot be differentiated based on the detected energy.

Methods: Recently, the authors have proposed a DIPET technique that uses a combination of radiotracer A which is a pure positron emitter (such as(18)F or (11)C) and radiotracer B in which positron decay is accompanied by the emission of a high-energy (HE) prompt gamma (such as (38)K or (60)Cu). Events that are detected as triple coincidences of HE gammas with the corresponding two 511 keV photons allow the authors to identify the lines-of-response (LORs) of isotope B. These LORs are used to separate the two intertwined distributions, using a dedicated image reconstruction algorithm. In this work the authors propose a new version of the DIPET EM-based reconstruction algorithm that allows the authors to include an additional, independent estimate of radiotracer A distribution which may be obtained if radioisotopes are administered using a staggered injections method. In this work the method is tested on simple simulations of static PET acquisitions.

Results: The authors' experiments performed using Monte-Carlo simulations with static acquisitions demonstrate that the combined method provides better results (crosstalk errors decrease by up to 50%) than the positron-gamma DIPET method or staggered injections alone.

Conclusions: The authors demonstrate that the authors' new EM algorithm which combines information from triple coincidences with prompt gammas and staggered injections improves the accuracy of DIPET reconstructions for static acquisitions so they reach almost the benchmark level calculated for perfectly separated tracers.

Figure 1

(a) A protocol for staggered/HE prompt gamma DIPET acquisition using radioisotope A (pure β⁺ emitter) and radioisotope B (that emits β⁺ and prompt HE gammas) using two static (nondynamic) acquisitions. The curves schematically represent the dynamics of the tracer. During the “A” phase of the scan a standard dual coincidence dataset ${\mathit{g}}^{{\mathit{A}}^{\prime }}$ is acquired, while during the “A and B” phase – two datasets: the dual coincidence dataset ${\mathit{g}}^{\mathit{A}B}$ and triple coincidence dataset ${\mathit{g}}^{{\mathit{B}}^{\prime }}$ are simultaneously acquired. (b) An example of the DIPET protocol where tracer A is imaged dynamically (note that the acquisition starts before the injection) and phase “A and B” is imaged statically. Any combination of static and dynamic acquisitions can be used. Please note, however, that only the first (static) acquisition type (a) was investigated in this paper and that different dynamic options, such as the one proposed in (b), will be studied in the future.

Figure 2

Phantoms used in GATE simulations. (a) Three spherical sources filled with radioisotope A = ¹⁸F and three spherical sources filled with B = ⁶⁰Cu. (b) Combination of nonoverlapping, partially overlapping, and fully overlapping sources A = ¹⁸F and B = ⁶⁰Cu. All sources were inserted into a 30 cm thick and 30 cm long cylinder filled with nonradioactive water. The percentage values indicate the fraction of some nominal activities of radioisotopes A and B filling each of the inserts.

Figure 3

Central transaxial slices of the first phantom images showing the distributions of radiotracers A (first row) and B (second row) reconstructed from six-sphere phantom [Fig. 2a]. The columns from left to right represent images: reconstructed from the benchmark data (Case #1), from the data acquired using staggered injections only (Case #2), using prompt gamma only (Case #3) and using staggered and prompt gamma combined (Case #4), respectively. “RMSE” indicates root mean square error estimated over the three ROIs corresponding to the location of the other radiotracer (the smaller is the value of RMSE, the more accurate is the method). Dashed lines indicate the location of another radiotracer. Arrows indicate the locations where the residual activity from other isotope is the most pronounced. Images are 128 × 128 × 128 with 3.9 mm voxels. Bottom of the figure: profiles drawn through the centers of upper two spheres for each case. The black line shows profiles drawn through images corresponding to tracer A (top row) and grey line to tracer B (bottom row).

Figure 4

Central transaxial slices of the second phantom images showing the distributions of radiotracers A (first row) and B (second row) reconstructed from overlapping phantom [Fig. 2b]. The columns from left to right represent images: reconstructed from the benchmark data (Case #1), from the data acquired using staggered injections only (Case #2), using prompt gamma only (Case #3), and using staggered and prompt gamma combined (Case #4), respectively. “RMSE” indicates root mean square error estimated over the three ROIs corresponding to the location of other radiotracer (the smaller is the value of RMSE, the more accurate is the method). Dashed lines indicate the location of another radiotracer. Arrows indicate the locations where the activity artifacts from other isotope are pronounced. Images are 128 × 128 × 128 with 3.9 mm voxels. Bottom of the figure: profiles drawn through the centers of upper three spheres for each case. The black line shows profiles drawn through images corresponding to tracer A (top row) and grey line to tracer B (bottom row).

Figure 5

The illustration of the effect of the number of iterations on image quality. The crosstalk between two radiotracers gradually decreases and disappears almost entirely by the 10th iteration. Central transaxial slices are displayed showing the images of radiotracers A (first row) and B (second row) reconstructed from six-sphere phantom [Fig. 2a] and using staggered and prompt gamma methods combined (case #4). Dashed lines indicate the location of another radiotracer. Images are 128 × 128 × 128 with 3.9 mm voxels. Bottom of the figure: profiles drawn through the centers of upper two spheres for each case. The black line shows profiles drawn through images corresponding to tracer A (top row) and grey line to tracer B (bottom row).