Myocardial perfusion imaging (MPI) is subjected to a variety of artifacts that can limit the performance of the study. Artifacts can arise at any stage in the MP process, and they can be grouped into issues related to the patient, equipment, or technologist [1].
Patient-related artifacts in MPI, not originated from the heart, are mainly due to attenuation, thoracic motions, and sub-diaphragmatic activity. Attenuation artifacts are caused due to the anatomy surrounding the heart [2]. Motion artifacts cause blurring along the motion direction, and they are mainly due to the cranio-caudal respiratory motion of the heart [3]. Sub-diaphragmatic organs adjacent to the heart, mainly the liver, present prominent radioactivity which mainly interferes with the adjacent inferior wall of the left ventricle due to photon scattering [4]. The radioactivity of the moving heart and liver, during the cranio-caudal respiratory motion, overlap and as a result it artificially decreases or increases the uptake in the adjacent regions of the heart. [5]. Different amplitudes of these motions may result in different spillover of the liver activity into the adjacent heart walls. The liver (a) activity, (b) proximity to the heart, and (c) motion amplitude during respiration are the main contributing parameters to MP artifacts due to liver [5]. This varying liver interaction causes a complex and clinically unpredictable variation in MPI which may mask cardiac defects in the case of an uptake increase, or cause artifactual defects in the case of an uptake decrease, or may create a complex combination of these two cases.
Phantoms that closely simulate tissues, anatomy and motion are important to investigate these artifacts and optimize the MP imaging technique, without exposing patients to radiation [6,7]. To investigate this varying liver interaction and its impact on defect detection, phantoms should be able to simulate various (a) cardiac-liver activity (CLA) ratios, (b) cardiac-liver proximities (CLP) and (c) oscillation amplitudes during respiration.
In single-photon emission computed tomography (SPECT) MPI, the Tc99m extra-cardiac activity (ECA) due to liver was studied using static and manually moving torso phantoms with static cardiac and liver compartments [4,5,8,9]. Germano et al. [4] investigated ECA, using a static torso phantom with an enclosed custom-made lucite wedged-shaped liver compartment for three different values of CLA. Nuyts et al. [8], using a static perspex cylinder with an enclosed plastic water-bag liver compartment, studied MP artifacts due to ECA for a CLA value. Heller et al. [9], using an almost similar phantom as Nuyts et al. [8], also studied MP artifacts due to ECA for a higher CLA value. Pitman et al. [5], using an acrylic cylinder with an enclosed static saline-bag liver compartment, examined the effects of ECA interaction for two different CLA values and two different diaphragmatic motion amplitudes by manually moving the cylinder during SPECT MP acquisitions.
In our previous studies [2,3,10–12], we acquired SPECT/CT images using an anthropomorphic thorax which enclosed moving thoracic compartments: (a) a cardiac phantom of an ECG beating and moving left ventricle during respiration in the cranio-caudal direction, and (b) a breathing phantom of a pair of lungs. These motions could be controlled in time interval of 0.1 sec and various parameters of the motions can vary such as the ejection fraction, heart rate, breath rate and tidal volume. In these studies, we characterized MP artifacts, evaluate physicians reports on cardiac defects, and correct the images for motion.
In this paper, we present the development of a liver phantom with motion, implemented within an anthropomorphic thorax and synchronized with the existing thoracic motions, and the evaluation of the phantom assembly in investigating MP artifacts due to liver.