Gillen, R; Erlandsson, K; Denis-Bacelar, A M; Thielemans, K; Hutton, B F; McQuaid, S J (2022) Towards accurate partial volume correction in 99mTc oncology SPECT - perturbation for case-specific resolution estimation. EJNMMI Physics, 9. 59

Preview

Text eid9601.pdf - Published Version Available under License Creative Commons Attribution. Download (3MB) | Preview

Abstract

Background: Currently, there is no consensus on the optimal Partial Volume Correction (PVC) algorithm for oncology imaging. Several existing PVC methods require knowledge of the reconstructed resolution, usually in the form of the point spread function (PSF), which is often assumed to be spatially invariant. However, this is not the case for SPECT imaging. This work aimed to assess the improvement in accuracy of SPECT quantification when PVC is applied using a PSF estimated by perturbation.

Methods: SPECT simulations were performed for a range of activity distributions, including those which replicated typical clinical oncology Tc-99m imaging studies. Perturbation using a small point-source was used to estimate Gaussian PSFs in reconstructed images. The PSF was measured for different positions in the field of view, for different lesion shapes, sizes and contrasts, and for noise-free and noisy data. The accuracy of the measured PSFs was assessed by comparing reconstructed images with ground truth images convolved with the measured PSF and calculating the Root Mean Square Error (RMSE). PVC was applied using the Single Target Correction (STC) method, incorporating the perturbation-based PSF and the corrected regional mean values (RMV) were assessed for quantitative accuracy by comparing with the ground truth.

Results: Perturbation-based PSF estimates demonstrated dependence on the position in the Field of View (FOV) and the number of OSEM iterations. A lower RMSE between reconstructed images and ground truth images convolved with the perturbation-estimated PSF was observed compared with performing convolution with a PSF reflecting the resolution at the centre of the FOV (i.e. not case-specific). RMVs following PVC using the perturbation-estimated PSF were more accurate than uncorrected data, or data corrected with PVC using the FOV-centre PSF. This was the case for both simple and anthropomorphic phantoms. For the simple phantom, RMVs were within 0.7% of the ground truth values. Accuracy improved after 5 or more OSEM iterations (10 subsets). For the anthropomorphic phantom, post-correction RMVs were within 1.6 % of the ground truth values for noise-free data.

Conclusion: Perturbation using a simulated point source could potentially improve quantitative SPECT accuracy via the application of PVC, provided that sufficient reconstruction iterations are used.

Item Type:	Article
Subjects:	Ionising Radiation > Dosimetry
Divisions:	Medical, Marine & Nuclear
Identification number/DOI:	10.1186/s40658-022-00489-5
Last Modified:	18 Nov 2022 14:35
URI:	https://eprintspublications.npl.co.uk/id/eprint/9601