Ejection fraction in myocardial perfusion imaging assessed with a dynamic phantom: comparison between IQ-SPECT and LEHR

Figures

• 

  Fig. 1. The dynamic heart phantom. Left—the membrane and pump units. Right—the phantom assembled and ready for measurement in the gamma camera. The membrane unit installed is shown inside a NEMA IEC PET body phantom and positioned for IQ-SPECT acquisition

• 

  Fig. 2. CT imaging was used in the phantom validation d. Segmentation between phantom wall and water was performed by fitting a double Gaussian function to the CT volume histogram a. The volume corresponding to the ESV is delineated in b and c by the thin black line including the inner membrane. The thick black line indicates the corresponding slice locations in b and c. The segmentation was verified by visually viewing the axial slices c

• 

  Fig. 3. The phantom volumes analysed from the gated CT data. The largest and smallest volume values were used to calculate the EF. All the shown 21 phase volumes obtained from the gated CT scans were analysed as an additional verification of the phantom and CT analysis method: the expected sinusoidal-like curves were produced for all the settings. The values include the innermost water volume and the phantom membrane surrounding it as shown in Fig. 2b, c

• 

  Fig. 4. SPECT images of the phantom. Representative slices of the phantom EF setting 65% acquired using LEHR and IQ-SPECT protocols. Images are at end-diastolic (EDV) and end-systolic (ESV) phases after 2, 12 and 34 equivalent iterations presented with corresponding EF values. The contours present QPS software volume delineations

• 

  Fig. 5. EFLEHR and EFIQSPECT estimates as a function of number of reconstruction equivalent iterations for the dynamic heart phantom, analysed using a QGS and b 4DM software. The phantom was imaged with 45, 55, 65 and 70% EF settings (grey solid line)

• 

  Fig. 6. ESV and EDV estimates of the dynamic heart phantom using LEHR and IQ-SPECT protocols and analysed using a QGS and b 4DM. The phantom results for phantom EF settings 45 and 70% are shown. The long and short dashed lines represent the EDVtrue and the ESVtrue volumes, respectively, determined from the gated CT images

• 

  Fig. 7. Cardiac phantom volume estimates for LEHR and IQ-SPECT protocols compared with corresponding CT measurements (small dot) with phantom EF setting 65%. The effect of the number of reconstruction iterations (EI) is shown with different markers (cross = 2, square = 12 and large dot = 34)