PT - JOURNAL ARTICLE AU - Frank DiFilippo AU - Aniruddha Doke AU - Yuqing Zhang AU - Steve Huang TI - Quantification of murine tumor uptake with clinical PET/CT and super-resolution DP - 2014 May 01 TA - Journal of Nuclear Medicine PG - 375--375 VI - 55 IP - supplement 1 4099 - http://jnm.snmjournals.org/content/55/supplement_1/375.short 4100 - http://jnm.snmjournals.org/content/55/supplement_1/375.full SO - J Nucl Med2014 May 01; 55 AB - 375 Objectives Although small animal PET/CT is designed for preclinical research, the wide availability of clinical PET/CT makes it a desirable alternative. A main limitation of clinical PET/CT for murine imaging is its PET spatial resolution (4.5 mm FWHM). Furthermore, clinical software for image reconstruction is not optimized for imaging mice. In this study we evaluate the feasibility of clinical PET/CT with super-resolution (SR) in a biodistribution study of a tumor-specific PET tracer in a xenograft mouse model. Methods PET data of mice were acquired on a clinical scanner (Biograph mCT) at eight stepped locations over a 4.8 mm cube for one minute per frame at two hours post injection. Each location's image was reconstructed using the clinical OS-EM software, and the SR image then was derived from the eight images using iterative deconvolution. Dynamic PET data also were acquired for nine 5-minute frames without stepping the mouse and were processed by iterative deconvolution. After imaging, ex-vivo measurements were made of tumor dimensions and of tumor activity in a well counter. Results SR images were free of artifact and had acceptable image quality and contrast for accurate quantitative analysis. Dynamic images had some residual artifact from the scanner’s coarse pixel size but produced usable time-activity curves. Tumors of 2.1 to 4.5 mm cross-sectional diameter and also the renal cortex were visualized clearly in both SR and dynamic images. Based on the dimensions of the excised tumors, recovery coefficients versus spatial resolution were calculated and compared to actual recovery coefficients derived from SR image and well count measurements. The measurements were consistent with an effective PET spatial resolution of 2.9 mm FWHM. Conclusions A clinical PET/CT scanner with SR acquisition and processing is feasible for tumor model studies in mice. Although image quality is not equivalent to that of state-of-the-art small animal PET/CT, it produces acceptable results for many preclinical research studies in mice.