Estimation of optimal number of gates in dual gated 18F-FDG cardiac PET

Abstract

Gating of positron emission tomography images has been shown to reduce the motion effects, especially when imaging small targets, such as coronary plaques. However, the selection of optimal number of gates for gating remains a challenge. Selecting too high number of gates results in a loss of signal-to-noise ratio, while too low number of gates does remove only part of the motion. Here, we introduce a respiratory-cardiac motion model to determine the optimal number of respiratory and cardiac gates. We evaluate the model using a realistic heart phantom and data from 12 cardiac patients (47-77 years, 64.5 on average). To demonstrate the benefits of our model, we compared it with an existing respiratory model. Based on our study, the optimal number of gates was determined to be five respiratory and four cardiac gates in the phantom and patient studies. In the phantom study, the diameter of the most active hot spot was reduced by 24% in the dual gated images compared to non-gated images. In the patient study, the thickness of myocardium wall was reduced on average by 21%. In conclusion, the motion model can be used for estimating the optimal number of respiratory and cardiac gates for dual gating.

Figure 1

Modelling and analysis workflow. The motion modelling consists of the following 4 steps. Step 1: Acquisition and dual gating of the PET scans with up to 60 gates. Step 2: Automatic motion detection and individual motion models for respiratory, cardiac and dual motion. Step 3: Individual and general models for number of gates in terms of respiratory, cardiac and dual motion. Step 4: Evaluation of the general model for optimal number of gates using SNR and line profiles. General models were built only for the patient study. Numbers in parenthesis refer to the models defined by the corresponding equations.

Figure 2

The optimal number of respiratory (a), cardiac (b) and dual (c) gates as a function of motion in mm (models (10), (11) and (12), respectively). Line is a linear fit of optimal number of gates (dots) as given in Table 6 and the motion given in Supplementary Data 2, describing the minimum number of gates needed for motion compensation in terms of the scanner resolution.

Figure 3

SNR for each gating scheme in the (a) phantom study and (b) patient study. The SNR is given as a function of respiratory gates where the individual line plots represent different cardiac gating schemes. The number of cardiac gates is 1, 4, 6, 8 and 10 for (a) and 1, 6 and 8 for (b).

Figure 4

SNR versus the number of gates in the phantom study (a) and in the patient study for patient 11 (b). Plus signs indicate measured data, circles show SNR with no gating and optimal number of gates based on model (4) as in Table, and the curve describes the fitted SNR by equation $a+b\ast {\log }_{10}(n)$.

Figure 5

Comparison of different gating schemes with (a) non-gated and dual gated PET images with (b) three respiratory and three cardiac gates, (c) five respiratory and four pulsatile gates and (d) six respiratory and five pulsatile gates in the phantom study. The hot spot is blurred in the non-gated image versus dual gated images. The hot spot in the phantom is slightly smaller in (c) and (d) whereas (d) shows slightly increased noise compared to (c).

Figure 6

Comparison of different gating schemes with (a) non-gated and dual gated PET images with (b) three respiratory and three cardiac gates, (c) five respiratory and four pulsatile gates and (d) six respiratory and five pulsatile gates in the patient study, subject 11. The myocardium is slightly better delineated in (c) and (d). The difference between (c) and (d) is very small.

Figure 7

Example profiles of the (a) phantom over the hotspot in slice number 25 and (b) on subject 12 over the myocardium in slice number 23. The advantage of using five respiratory gates and four cardiac gates over three respiratory and three cardiac gates is better delineation of the hot spot profile and myocardium. However, increasing the number of gates over five respiratory and four cardiac gates does not result in significant increase in the profile peak.

Figure 8

Linear functions for the optimal number of gates as a function of the amount of motion. Red solid line is fit for Dawood’s model as in (14), and blue dashed line is for our model as in (10). Both functions show similar behaviour over the motion range in this study.