Quantification of myocardium at risk in myocardial perfusion SPECT by co-registration and fusion with delayed contrast-enhanced magnetic resonance imaging--an experimental ex vivo study.

BACKGROUND: Myocardial perfusion single-photon emission computed tomography (MPS) can be used to assess myocardium at risk in occlusive coronary ischaemia. The aim was to develop a method to quantify myocardium at risk as perfusion defect size on ex vivo MPS using co-registration and fusion with ex vivo magnetic resonance imaging (MRI). METHODS: Pigs (n = 19) were injected 99mTc-tetrofosmin prior to concluding 40 min of coronary artery occlusion, followed by reperfusion and MRI contrast injection. The excised heart was imaged with T1-weighted MRI and MPS, and images were co-registered using freely available software (Segment v1.8, http://segment.heiberg.se). The left ventricle was semi-automatically delineated in MRI and copied to MPS. The threshold for a MPS perfusion defect was defined as the mean counts in the MPS image at the MRI-determined border between remote myocardium and air. The threshold was measured using count maxima set to the 100th-95th percentile of counts within the myocardium. The count maximum that gave the lowest threshold variability (SD) was considered the most robust. RESULTS: A count maximum using the 100th percentile yielded a threshold of (mean ± SD) 55 ± 6·2%. This method showed the lowest SD compared to 99th-95th percentile count maxima (6·6-7·2%). CONCLUSIONS: We describe a method for objective quantification of myocardium at risk as perfusion defect size on MPS using knowledge of the anatomy of the myocardium from co-registered MRI. This enables simultaneous quantification of myocardium at risk by MPS and infarct size by MRI for the evaluation of treatments for myocardial infarction.

Development and validation of a new automatic algorithm for quantification of left ventricular volumes and function in gated myocardial perfusion SPECT using cardiac magnetic resonance as reference standard.

BACKGROUND: By gating image acquisition in myocardial perfusion SPECT (MPS) to ECG, left ventricular (LV) volumes and function can be determined. Several previous studies have shown that existing MPS software packages underestimate LV volumes compared to cardiac magnetic resonance (CMR). The aim of this study was therefore to develop a new LV segmentation algorithm for gated MPS using CMR as reference standard. METHODS AND RESULTS: A total of 126 patients with suspected coronary artery disease, who underwent both gated MPS and CMR were retrospectively included. The proposed LV segmentation algorithm (Segment) was trained in 26 patients, and tested in 100 patients in comparison to four commercially available MPS software packages (QGS, MyoMetrix, ECTb, and Exini) using CMR as reference standard. Mean bias ± SD between MPS and CMR was for EDV -5% ± 12%, -43% ± 8%, -40% ± 8%, -42% ± 9%, -32% ± 7%, for ESV 0% ± 17%, -41% ± 16%, -34% ± 15%, -54% ± 13%, -41% ± 10%, for EF -2% ± 13%, -1% ± 14%, -7% ± 15%, 17% ± 16%, 10% ± 17% for Segment, QGS, MyoMetrix, ECTb, and Exini, respectively, and for LVM 3% ± 18%, 33% ± 25%, 37% ± 24% for Segment, QGS, and ECTb, respectively. Correlation between MPS by Segment and CMR were for EDV R (2) = 0.89, for ESV R (2) = 0.92, for EF R (2) = 0.69, and for LVM R (2) = 0.72, with no difference compared to the correlation between the other MPS software packages and CMR (EDV R (2) = 0.86-0.92, ESV R (2) = 0.91-0.93, EF R (2) = 0.64-0.65, and LVM R (2) = 0.68-0.70). CONCLUSION: The Segment software quantifies LV volumes and EF by MPS with similar correlation and a low bias compared to other MPS software packages, using CMR as reference standard. Hence, the Segment software shows potential to provide clinically relevant volumes and functional values from MPS.

An automatic method for quantification of myocardium at risk from myocardial perfusion SPECT in patients with acute coronary occlusion.

BACKGROUND: In order to determine myocardial salvage, accurate quantification of myocardium at risk (MaR) is necessary. We present a validated novel automatic segmentation algorithm for quantification of MaR by myocardial perfusion SPECT (MPS) in patients with acute coronary occlusion. METHODS AND RESULTS: Twenty-nine patients with coronary occlusion were injected with a perfusion tracer before reperfusion, and underwent rest MPS within 4 hours. The MaR was quantified using the proposed algorithm (Segment software), the software Quantitative Perfusion SPECT (QPS) and by manual segmentation. The Segment MaR algorithm used a threshold of 55% of maximal counts and an a priori model based on normal coronary artery perfusion territories. The MaR was 30 ± 10% left ventricular mass (%LVM) by manual segmentation, 31 ± 12%LVM by Segment, and 36 ± 14%LVM by QPS. There was a good agreement between automatic and manual segmentation for both of the algorithms with a lower bias for Segment (.8 ± 4.0%LVM) than for QPS (5.8 ± 5.8%LVM) when compared to manual segmentation. CONCLUSIONS: The Segment MaR algorithm can be used to correctly assess MaR from MPS images in patients with acute coronary occlusion without access to tracer-specific normal database. The MaR in relation to final infarct size enables determination of myocardial salvage.

An improved method for automatic segmentation of the left ventricle in myocardial perfusion SPECT.

UNLABELLED: This study describes and validates a new method for automatic segmentation of left ventricular mass (LVM) in myocardial perfusion SPECT (MPS) images. This is important for estimating the size of a perfusion defect as percentage of the left ventricle. METHODS: A total of 101 patients with known or suspected coronary artery disease underwent both rest and stress MPS and MRI. A new automated algorithm was trained in 20 patients (40 MPS studies) and tested in 81 patients (162 MPS studies). The algorithm, which segmented the left ventricle in the MPS images, is based on Dijkstra's algorithm and finds an optimal mid-mural line through the left ventricular wall. From this line, the endocardium and epicardium are identified on the basis of an individually estimated wall thickness and signal intensity. The algorithm was validated by comparing LVM in both stress and rest MPS, with LVM of the manually segmented left ventricle from MRI as the reference standard. For comparison, LVM was quantified using the software quantitative perfusion SPECT (QPS). RESULTS: The mean difference+/-SD in LVM between MPS and MRI was lower for the new method (6%+/-15% LVM) than for QPS (18%+/-19% LVM) for both mean difference (P<0.001) and SD (P=0.015). Linear regression analysis of LVM, comparing MPS and MRI, yielded R2=0.83 using the new method and R2=0.80 using QPS. Interstudy variability, measured as the coefficient of variance between rest MPS and stress MPS, was 6% for both the new method and QPS. Both the new algorithm and QPS systematically overestimated LVM in hearts with thin myocardium and underestimated LVM in hearts with thick myocardium. CONCLUSION: The new segmentation algorithm quantifies LVM with a significantly lower bias and variability than does the commercially available QPS software, when compared to manually segmented LVM by MRI. This makes the new algorithm an attractive method to use for estimating the size of the perfusion defect when expressing it as percentage of the left ventricle. This study shows that inaccurate estimation of wall thickness is the main source of error in automatic segmentation.