Phantom and clinical evaluation of bone SPECT/CT image reconstruction with xSPECT algorithm

Area of Science:

• Nuclear Medicine
• Medical Imaging
• Radiology

Background:

• Novel image reconstruction methods, xSPECT Quant (xQ) and xSPECT Bone (xB), utilize ordered subset conjugate gradient minimizer (OSCGM) for SPECT/CT.
• Conventional Flash3D (F3D) reconstruction is widely used for SPECT/CT imaging.

Purpose of the Study:

• To compare the performance characteristics of xQ, xB, and F3D reconstruction algorithms.
• To evaluate image quality metrics including spatial resolution, contrast-to-noise ratio, and recovery coefficients.

Main Methods:

• Phantom studies using custom-designed body and spine phantoms with varying activity concentrations and tumor-to-normal bone ratios (TBR).
• Evaluation of full width at half maximum, coefficients of variance (CV), contrast-to-noise ratio (CNR), and recovery coefficients (RC).
• Retrospective analysis of 20 patient SPECT/CT datasets for bone metastases using xQ and xB reconstruction.

Main Results:

• xSPECT Bone (xB) demonstrated superior spatial resolution compared to xQ and F3D.
• xQ and xB showed improved CV and RC over F3D.
• CNR peaked at different iterations for xQ (24) and F3D/xB (48).
• Significant differences in SUVmax and SUVpeak were observed between xQ and xB in patient data.

Conclusions:

• xSPECT Quant (xQ) and xSPECT Bone (xB) provide more accurate SPECT/CT reconstructions than F3D.
• xB offers consistent spatial resolution in bone SPECT imaging, independent of iteration convergence.
• xB reconstruction enhances SPECT image quality by integrating CT data, improving diagnostic insights.