Algorithm delivers quantitative F-18 FDG PET partial volume correction

The use of algorithms is changing the game for close-as-possible assessment of tumor volume and resolution recovery from PET cancer imaging, especially small objects affected by partial volume effects. Researchers conducting phantom studies have optimized quantitative tumor delineation of F-18 FDG PET imaging with two specialized algorithms that provide accurate partial volume correction for volumes as small as one-third of a milliliter, according to research published May 8 in the Journal of Nuclear Medicine.

Elin Wallsten, PhD, from the department of radiation sciences at Umea University in Umea, Sweden, and colleagues conducted phantom imaging with balloons for tumor volume delineation and partial-volume correction in FDG PET using a combination of two algorithms for interative reconstruction, with one dedicated to resolution recovery. Results showed that these combined software protocols accurately corrected partial volume effects and estimated tumor volume to 0.37 mL.  

“The aims of this study were to investigate a slightly modified version of a method for the delineation of tumor volume and to correct the measured mean activity within the volume for [partial volume effect (PVE)] using correction factors,” wrote Wallsten et al. “The aim was to study the effect on small volumes such as lymph nodes and small tumors, and it was therefore of particular interest to evaluate possible improvements when including a modern resolution recovery algorithm in the reconstruction.”

Five small balloons of varying shapes and volumes of 0.1 to 0.7 mL filled with F-18 FDG were used to test the algorithms. Tumor volumes were delineated by looking at the weighted sum of background and tumor intensities, and the threshold intensity and volumes of interest were then used to determine corrective measures via system point-spread function and tumor masking. Image reconstruction was achieved with two iterations, 24 subsets and a Gaussian-type data filter for post-processing of 6.4 mm. The resolution recovery algorithm included three iterations, 24 subsets and the same filter but of 3 mm in full width at half maximum. Resulting image data were compared with the uncorrected mean uptake value and maximum uptake volume.

“The PVE-corrected values in this study are generally more accurate than the maximum values for [volumes of interest] and have less statistical variation, thus making this the superior method,” concluded the authors. “Statistical variation is increased, compared with uncorrected values, due to uncertainties in volume delineation, but the mean is substantially more accurate.”

Additionally, lesion activity was effectively quantified using the dual-algorithm method and data reconstruction without resolution recovery were not viable for assessing activity or for tumor delineation. “The results give further support for implementation of resolution recovery methods in PET imaging,” wrote the authors.