Evaluation of a decision support system for interpretation of myocardial perfusion gated SPECT.

PURPOSE: We have recently presented a decision support system for interpreting myocardial perfusion scintigraphy (MPS). In this study, we wanted to evaluate the system in a separate hospital from where it was trained and to compare it with a quantification software package. METHODS: A completely automated method based on neural networks was trained for the interpretation of MPS regarding myocardial ischaemia and infarction using 418 MPS from one hospital. Features from each examination describing rest and stress perfusion, regional and global function were used as inputs to different neural networks. After the training session, the system was evaluated using 532 MPS from another hospital. The test images were also processed with the quantification software package Emory Cardiac Toolbox (ECTb). The images were interpreted by experienced clinicians at both the training and the test hospital, regarding the presence or absence of myocardial ischaemia and/or infarction and these interpretations were used as gold standard. RESULTS: The neural network showed a sensitivity of 90% and a specificity of 85% for myocardial ischaemia. The specificity for the ECTb was 46% (p < 0.001), measured at the same sensitivity. The neural network sensitivity for myocardial infarction was 89% and the specificity 96%. The corresponding specificity for the ECTb was 54% (p < 0.001). CONCLUSION: A decision support system based on neural networks presents interpretations more similar to experienced clinicians compared to a conventional automated quantification software package. This study shows the feasibility of disseminating the expertise of experienced clinicians to less experienced physicians by the use of neural networks.

The added value of ECG-gating for the diagnosis of myocardial infarction using myocardial perfusion scintigraphy and artificial neural networks.

To assess the value of ECG-gating for the diagnosis of myocardial infarction using myocardial perfusion scintigraphy (MPS) and an artificial neural network. A total of 422 patients referred for MPS were studied using a one day (99m)Tc-tetrofosmin protocol. Adenosine stress combined with submaximal dynamic exercise was used. The images were interpreted by one of three experienced clinicians and these interpretations regarding the presence or absence of myocardial infarction were used as the standard. A fully automated method using artificial neural networks was compared with the clinical interpretation. Either perfusion data alone or a combination of perfusion and function from ECG-gated images were used as input to different artificial neural networks. After a training session, the two types of neural networks were evaluated in separate test groups using an eightfold cross-validation procedure. The neural networks trained with both perfusion and ECG-gated images had a 4-7% higher specificity compared with the corresponding networks using perfusion data only, in four of five segments compared at the same level of sensitivity. The greatest improvement in specificity, from 70% to 77%, was seen in the inferior segment. In the septal and lateral segments the specificity rose from 73% to 77% and from 81% to 85%, respectively. In the anterior segment, the increase in specificity from 93% to 94% by adding functional data was not significant. The addition of functional information from ECG-gated MPS is of value for the diagnosis of myocardial infarction using an automated method of interpreting myocardial perfusion images.

Left ventricular function parameters obtained from gated myocardial perfusion SPECT imaging: a comparison of two data processing systems.

BACKGROUND AND AIM: The Cedars-Sinai Quantitative Gated Single Photon Emission Computed Tomography (SPECT) (QGS) program, used to quantify left ventricular function parameters from gated myocardial perfusion scintigraphy (MPS), has been extensively validated and compared with other methods of quantification. However, little is known about the reproducibility of QGS on different processing systems. This study compared the findings of QGS running on workstations provided by two different manufacturers. METHODS: Gated rest MPS studies of 50 patients were analysed retrospectively. Filtered back-projection (FBP) was performed using identical parameters on Philips Pegasys and Nuclear Diagnostics Hermes workstations to produce gated short-axis (SA) slices. In addition, the gated SA slices reconstructed on the Pegasys were transferred to the Hermes. QGS was used to calculate the end-diastolic volume (EDV), end-systolic volume (ESV) and left ventricular ejection fraction (LVEF) in each case. RESULTS: The mean+/-standard deviation differences between the Pegasys and Hermes function parameters were -7.06+/-3.91 ml (EDV), -5.54+/-3.21 ml (ESV) and +1.14%+/-1.43% (LVEF) when data were reconstructed on different systems, and -0.16+/-1.58 ml (EDV), -0.10+/-1.02 ml (ESV) and +0.14%+/-0.73% (LVEF) when data were reconstructed on the same system. Bland-Altman plots showed definite trends for EDV and ESV for data reconstructed on different systems, but no trends were seen for data reconstructed on the same system. CONCLUSIONS: When data were reconstructed on two separate systems, the difference between the function parameters obtained from Pegasys and Hermes could be ascribed to differences in the reconstruction process on each system despite the use of identical parameters (filters, etc). However, when the same reconstructed data were analysed on both systems, no significant difference in left ventricular function parameters was observed.

Enhanced myocardial contractility associated with hypercalcaemia of hyperparathyroidism - a case study in six patients.

The possible association of increased left ventricular ejection fraction (LVEF) in patients with increased serum Ca(2+) was observed in our clinic. Six patients with confirmed primary hyperparathyroidism and hypercalcaemia were studied prospectively. Tc-99m sestamibi gated SPECT was done pre- and postoperatively. The LVEF was abnormally high in all the patients pre-operatively, i.e. above the normal reference range (47-61%) as used in our clinic. It decreased in all of them postoperatively, yet in only three patients to values within the normal range. This was associated with normalisation of the serum Ca(2+) values. The median of the pre-operative LVEF was 74% and postoperative it was 61.5%. The median difference was 9% with a 95% CI for the median difference for paired data (6; 26). This was statistically significant. Increased LVEF was not previously described as part of the clinical picture of primary hyperparathyroidism. The in vivo effect of chronic hypercalcaemia on LV pump function my need to be revisited.