Technical Note: Virtual CT for reducing CT dose in targeted radionuclide therapy dosimetry

The effects of mismatch between SPECT and CT images on quantitative activity estimation - A simulation study

BACKGROUND

Quantitative activity estimation is essential in nuclear medicine imaging. Mismatch between SPECT and CT images at the same imaging time point due to patient movement degrades accuracy in both diagnostic studies and target radionuclide therapy dosimetry. This work aims to study the mismatch effects between CT and SPECT data on attenuation correction (AC), volume-of-interest (VOI) delineation, and registration for activity estimation.

METHODS

Nine 4D XCAT phantoms were generated at 1, 24, and 144 h post In-111 Zevalin injection, varying in activity distributions, body sizes, and organ sizes. Realistic noisy SPECT projections were generated by an analytical projector and reconstructed with a quantitative OS-EM method. CT images were shifted, corresponding to SPECT images at each imaging time point, from -5 to 5 voxels and also according to a clinical reference. The effect of mismatched AC maps was evaluated using mismatched CT images for AC in SPECT reconstruction while VOIs were mapped out from matched CTs. The effect of mismatched VOI drawings was evaluated using mismatched CTs to map out target organs while using matched CTs for AC. The effect of mismatched CT images for registration was evaluated by registering sequential mismatched CTs to align corresponding SPECT images, with no AC and VOI mismatch. Bi-exponential curve fitting was performed to obtain time-integrated activity (TIA). Organ activity errors (%OAE) and TIA errors (%TIAE) were calculated.

RESULTS

According to the clinical reference, %OAE was larger for organs near ribs for AC effect. For VOI effect, %OAE was larger for small and low uptake organs. For registration effect, %TIAE were larger when mismatch existed in more numbers of SPECT/CT images, while no substantial difference was observed when using mismatched CT at different imaging time points as registration reference. %TIAE was highest for VOI, followed by registration and AC, e.g., 20.62%±8.61%, 9.33%±4.66% and 1.13%±0.90% respectively for kidneys.

CONCLUSIONS

The mismatch between CT and SPECT images poses a significant impact on the accuracy of quantitative activity estimation, attributed particularly from VOI delineation errors. It is recommended to perform registration between emission and transmission images at the same time point to ensure diagnostic and dosimetric accuracy.

BIGDOSE: software for 3D personalized targeted radionuclide therapy dosimetry

Background

Advance 3D quantitative radionuclide imaging techniques boost the accuracy of targeted radionuclide therapy (TRT) dosimetry to voxel level. The goal of this work is to develop a comprehensive 3D dosimetric software, BIGDOSE, with new features of image registration and virtual CT for patient-specific dosimetry.

Methods

BIGDOSE includes a portable graphical user interface written in Python, integrating (I) input of sequential ECT/CT images; (II) segmentation; (III) non-rigid image registration; (IV) curve fitting and voxel-based integration; (V) dose conversion and (VI) 3D dose analysis. The accuracy of the software was evaluated using a simulation study with 9 XCAT phantoms. We simulated SPECT/CT acquisitions at 1, 12, 24, 72 and 144-hrs post In-111 Zevalin injection with inter-scans misalignments using an analytical projector for medium energy general purpose (MEGP) collimator, modeling attenuation, scatter and collimator-detector response. The SPECT data were reconstructed using quantitative OS-EM method. A CT organ-based registration was performed before the dose calculation. Organ absorbed doses for the corresponding Y-90 therapeutic agent were calculated on target organs and compared with those obtained from OLINDA/EXM, using dose measured from GATE as the gold standard. One patient with In-111 DTPAOC injection as well as two patients with Y-90 microsphere embolization were used to demonstrate the clinical effectiveness of our software.

Results

In the simulation, the organ dose errors of BIGDOSE were -9.59%±9.06%, -8.36±5.82%, -23.41%±6.67%, -6.05%±2.06% for liver, spleen, kidneys and lungs, while they were -25.72%±12.52%, -14.93%±10.91%, -28.63%±12.97% and -45.30%±5.84% for OLINDA/EXM. Cumulative dose volume histograms, dose maps and iso-dose contours provided 3D dose distribution information on the simulated and patient data.

Conclusions

BIGDOSE provides a one-stop platform for voxel-based dose estimation with enhanced functions. It is a promising tool to streamline the current clinical TRT dosimetric practice with high accuracy, incorporating 3D personalized imaging information for improved treatment outcome.