Derivation of a respiration trigger signal in small animal list-mode PET based on respiration-induced variations of the ECG signal

Todica, Andrei; Lehner, Sebastian; Wang, Hao; Zacherl, Mathias J.; Nekolla, Katharina; Mille, Erik; Xiong, Guoming; Bartenstein, Peter; Fougère, Christian la; Hacker, Marcus und Böning, Guido (2016): Derivation of a respiration trigger signal in small animal list-mode PET based on respiration-induced variations of the ECG signal. In: Journal of Nuclear Cardiology, Bd. 23, Nr. 1: S. 73-83

Abstract

Background. Raw PET list-mode data contains motion artifacts causing image blurring and decreased spatial resolution. Unless corrected, this leads to underestimation of the tracer uptake and overestimation of the lesion size, as well as inaccuracies with regard to left ventricular volume and ejection fraction (LVEF), especially in small animal imaging. Methods and Results. A respiratory trigger signal from respiration-induced variations in the electro-cardiogram (ECG) was detected. Original and revised list-mode PET data were used for calculation of left ventricular function parameters using both respiratory gating techniques. For adequately triggered datasets we saw no difference in mean respiratory cycle period between the reference standard (RRS) and the ECG-based (ERS) methods (1120 +/- 159 ms vs 1120 +/- 159 ms;P = n.s.). While the ECG-based method showed somewhat higher signal noise (66 +/- 22 ms vs 51 +/- 29 ms;P<.001), both respiratory triggering techniques yielded similar estimates for EDV, ESV, LVEF (RRS: 387 +/- 56 mu L, 162 +/- 34 mu L, 59 +/- 5%;ERS: 389 +/- 59 mu L, 163 +/- 35 mu L, 59 +/- 4%;P = n.s.). Conclusions. This study showed that respiratory gating signals can be accurately derived from cardiac trigger information alone, without the additional requirement for dedicated measurement of the respiratory motion in rats.

• Dokumententyp: Zeitschriftenartikel
• Fakultät: Medizin
• Themengebiete: 600 Technik, Medizin, angewandte Wissenschaften > 610 Medizin und Gesundheit
• ISSN: 1071-3581
• Sprache: Englisch