HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text Full Text (PDF) A correction has been published Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Itti, E. Articles by Meignan, M. Search for Related Content PubMed PubMed Citation Articles by Itti, E. Articles by Meignan, M.

Myocardial Tracking, A New Method to Calculate Ejection Fraction with Gated SPECT: Validation with ²⁰¹Tl Versus Planar Angiography

Department of Nuclear Medicine, Henri Mondor Hospital, Paris XII University, Créteil; HealthCenter Internet Services, Sophia-Antipolis, France; and Department of Nuclear Medicine, Military Hospital of Tunis, Tunisia

View larger version (46K): [in a new window]   FIGURE 1. Example of LVEF processing within same patient with QGS (A) and MTK (B) methods. Note difference in myocardial edge detection (arrowheads), which may alter left-ventricular–time curve (arrows) between QGS (C) and MTK (D). Ejection fraction was measured at 45% using ERNA, 41% using MTK, and 30% using QGS.

View larger version (68K): [in a new window]   FIGURE 2. Description of MTK algorithm for calculation of ejection fraction. Unique myocardial segmentation is performed on first time-frame volume (1, end-diastole) with matching to reference template to obtain morphological constraints of perfusion distribution, particularly for determination of valve plane. This procedure generates two 3D elastic ellipsoids (2, endocardial surface is represented mapped with perfusion polar map and epicardial surface is represented as wire grid). Node positions of these ellipsoids are tracked throughout 8 frames of cardiac cycle (3).

View larger version (39K): [in a new window]   FIGURE 3. Bland–Altman representation of accuracy of gated SPECT ejection fractions compared with planar ERNA, calculated with QGS at 20 min (A) and at 4 h (B) and calculated with MTK at 20 min (C) and at 4 h (D). Except for MTK at 20 min, each process led to significant underestimation.

View larger version (29K): [in a new window]   FIGURE 4. After dividing population into 5 subgroups according to ERNA range, we show that underestimation of LVEF by QGS at 20 min (A) and at 4 h (B) progressively increases when actual LVEF is high. Paired t tests are significant in groups with LVEF 30%. MTK at 20 min (C) accurately measures LVEF with no significant differences between gated SPECT and ERNA values in all subgroups. Results obtained with MTK at 4 h (D) show similar progressive underestimation of LVEF 40%, as with QGS. Boxes represent mean underestimations with associated error bars.

View larger version (12K): [in a new window]   FIGURE 5. Low reconstruction zooms artificially underestimate LVEF calculated with MTK. This systematic error can be avoided by using zoom factor of 2.0 or higher.

View larger version (17K): [in a new window]   FIGURE 6. Regression analysis between LVEF calculated from repeated gated SPECT acquisitions (early and delayed) shows good reproducibility of Germano’s algorithm despite significant count decay. By contrast, LVEF values measured during delayed/low-count acquisitions with MTK are significantly lower than those measured during early/high-count acquisitions.