Quantification of Partial Volume Effects in salivary glands SPECT images acquired after radiation therapy of head and neck tumours.

Document Type: Original Paper


Department of Medical Physics Sefako Makgatho Health Sciences University SouthAfrica


Introduction: Radical radiation therapy of head and neck cancers may injure salivary glands leading to reduced function. Single Photon Emission Computed Tomography (SPECT) images are used to quantify function. However, accurate quantification is hindered by partial volume effects (PVEs). The PVEs spread activity counts outside the glands hence under estimation of image counts. This study introduces a PVEs quantification technique aimed at recovering image counts spread outside salivary glands images.
Methods: The left and right parotid and submandibular salivary glands, were mimicked with hollow spheres LP, RP, LSM and RSM (diameter 16 mm; 14 mm; 11 mm and 12 mm) respectively. Technetium-99m solution (activity concentration 300 kBq/mL) filled the salivary glands prior to implanting into a hollow head and neck phantom later filled with technetium-99m solution (activity concentration 1440 Bq/mL). A SPECT image was acquired on 128 × 128 matrix size over 30 minutes and then reconstructed using a filtered backprojection algorithm utilizing a Butterworth filter with cut-off frequency 0.9 cycles per pixel and an order of 9. Reconstructed images were quantified using ImageJ software.
Results: The image counts extracted from the LP, RP, RMS and LMS salivary glands SPECT images were 672 019; 494 842; 398 091 and 262 908 respectively after quantification of PVEs compared to 486 320; 347 534, 272 940 and 175 307 before PVEs quantification. The established respective quantitative errors were: 27%; 29%; 31%, and 33%.
Conclusion: Quantification of PVEs in salivary gland SPECT imaging allows recovery of images counts spread outside the image pixels leading to accurate quantification of function.


Main Subjects

Articles in Press, Accepted Manuscript
Available Online from 01 March 2019
  • Receive Date: 13 November 2018
  • Revise Date: 26 February 2019
  • Accept Date: 01 March 2019