Respiratory tracking using EDR for list-mode binning in cardiac emission tomography: Comparison with MRI heart motion measurements

Abstract

In cardiac emission tomography, the respiratory motion of the heart forms an important source of artifact in the reconstructed images, which can mislead diagnosis. Various methods have been developed to measure the respiratory signal externally (e.g., bellow and visual tracking systems) and estimate the motion of the heart. However, most of these methods tend to be time consuming, costly and require additional effort both from patients and technicians. An alternative to these methods is ECG/EMG derived respiration (EDR), which uses ECG leads already placed on the patient without requiring any additional hardware set-up. In this work, we performed a series of volunteer MRI studies to compare the actual heart motion and EDR signals. Simultaneous to these acquisitions, external motion information from a bellow placed on the abdomen and a visual tracking system (VTS) were also obtained. These external measurements were compared with the "true" respiratory heart motion information provided by MRI. Our preliminary work indicated that the EDR-in average-had lower correlation values compared to the other two external methods. However, this must be mainly because of the ECG distortion in the MR magnetic field. As the EDR does not require any additional instrument, it could become a practical respiratory tracking method for cardiac emission tomography provided the issues with body motion are addressed.