Martin T. King MD, PhD, Presenter: Nothing to Disclose

Colin Carpenter, Abstract Co-Author: CEO, Siris Medical Systems

Conroy Sun, Abstract Co-Author: Nothing to Disclose

Xiaowei Ma, Abstract Co-Author: Nothing to Disclose

Guillem Pratx, Abstract Co-Author: Nothing to Disclose

Lei Xing PhD, Abstract Co-Author: Research Grant, Varian Medical Systems, Inc

Although Cerenkov luminescence imaging (CLI) is a validated method for imaging FDG-avid tumors, CLI suffers from low sensitivity. CCD-based beta imaging is an alternative modality, which incorporates a scintillator for improving sensitivity. In this abstract, we conduct a comparative evaluation of beta imaging with CLI.

We performed all experiments on a commercial small animal imager. For the scintillator, we used a radioisotopic screen. Each beta image was calculated as the difference between two images: 1) a ‘gamma + beta’ image with a scintillator between the object and the camera, and 2) a ‘gamma’ image with a 1 mm steel slab (>97.4% beta rejection from Tl-204) between the scintillator and the object. Exposure times for CLI and beta images were 180 s and 10-30 s, respectively. For in vitro FDG experiments, we obtained CLI and beta images of the following: a uniform source (100 uCi), a resolution phantom (1.2-4.8 mm rods), a 1.0 mm diameter capillary tube (100 uCi), and serial 10-fold dilutions of FDG droplets (10 uCi to 0.001 uCi ) in 50 uL matrigel. For in-vivo experiments, we imaged 5 nude mice inoculated with B16F10-luciferase expressing mouse melanoma cells. After excising the skin, we obtained a beta image, a CLI image, and a bioluminescence (BLI) image. We computed signal-to-background ratios (SBRs) for each modality. We then applied a dependent t-test to compare the SBRs between CLI and beta images.

For the uniform source, beta imaging provided 227 fold more counts-per-second than CLI. Beta imaging could not resolve any rods on the resolution phantom, whereas CLI could resolve rods at all diameters. For the capillary source, full-width-at-half-maximums for the beta and CLI images were 3.9 and 1.0 mm, respectively. For the FDG dilutions, beta imaging and CLI could detect 0.01 uCi and 0.1 uCi droplets, respectively, at SBRs greater than 1.5. For the in-vivo experiments, the average SBR was 16.1 +/- 11.2 for BLI, 2.1 +/- 1.6 for CLI, and 4.3 +/- 2.7 for beta images. SBRs for CLI and beta images were statistically different on dependent t-test (p = 0.05).

CCD-based beta imaging is more sensitive than CLI in both in vitro and in vivo models, but exhibits decreased spatial resolution.

CCD-based beta imaging can image FDG-avid tumors in a clinically feasible timeframe. This modality potentially may be used for FDG-guided surgery or endoscopy.

King, M, Carpenter, C, Sun, C, Ma, X, Pratx, G, Xing, L, A Comparative Evaluation of CCD-based Beta Imaging and Cerenkov Luminescence Imaging.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14015718.html