Assessment of absolute Tc-99m tetrofosmin retention in the myocardium as an index of myocardial blood flow and coronary flow reserve by gated-SPECT/CT: a feasibility study

Multi-input deep learning approach for Cardiovascular Disease diagnosis using Myocardial Perfusion Imaging and clinical data

PURPOSE

Accurate detection and treatment of Coronary Artery Disease is mainly based on invasive Coronary Angiography, which could be avoided provided that a robust, non-invasive detection methodology emerged. Despite the progress of computational systems, this remains a challenging issue. The present research investigates Machine Learning and Deep Learning methods in competing with the medical experts' diagnostic yield. Although the highly accurate detection of Coronary Artery Disease, even from the experts, is presently implausible, developing Artificial Intelligence models to compete with the human eye and expertise is the first step towards a state-of-the-art Computer-Aided Diagnostic system.

METHODS

A set of 566 patient samples is analysed. The dataset contains Polar Maps derived from scintigraphic Myocardial Perfusion Imaging studies, clinical data, and Coronary Angiography results. The latter is considered as reference standard. For the classification of the medical images, the InceptionV3 Convolutional Neural Network is employed, while, for the categorical and continuous features, Neural Networks and Random Forest classifier are proposed.

RESULTS

The research suggests that an optimal strategy competing with the medical expert's accuracy involves a hybrid multi-input network composed of InceptionV3 and a Random Forest. This method matches the expert's accuracy, which is 79.15% in the particular dataset.

CONCLUSION

Image classification using deep learning methods can cooperate with clinical data classification methods to enhance the robustness of the predicting model, aiming to compete with the medical expert's ability to identify Coronary Artery Disease subjects, from a large scale patient dataset.

Recent advances in cardiac SPECT instrumentation and imaging methods

Cardiac SPECT continues to play a critical role in detecting and managing cardiovascular disease, in particularly coronary artery disease (CAD) (Jaarsma et al 2012 J. Am. Coll. Cardiol. 59 1719-28), (Agostini et al 2016 Eur. J. Nucl. Med. Mol. Imaging 43 2423-32). While conventional dual-head SPECT scanners using parallel-hole collimators and scintillation crystals with photomultiplier tubes are still the workhorse of cardiac SPECT, they have the limitations of low photon sensitivity (~130 count s^-1 MBq^-1), poor image resolution (~15 mm) (Imbert et al 2012 J. Nucl. Med. 53 1897-903), relatively long acquisition time, inefficient use of the detector, high radiation dose, etc. Recently our field observed an exciting growth of new developments of dedicated cardiac scanners and collimators, as well as novel imaging algorithms for quantitative cardiac SPECT. These developments have opened doors to new applications with potential clinical impact, including ultra-low-dose imaging, absolute quantification of myocardial blood flow (MBF) and coronary flow reserve (CFR), multi-radionuclide imaging, and improved image quality as a result of attenuation, scatter, motion, and partial volume corrections (PVCs). In this article, we review the recent advances in cardiac SPECT instrumentation and imaging methods. This review mainly focuses on the most recent developments published since 2012 and points to the future of cardiac SPECT from an imaging physics perspective.

Noninvasive Nuclear SPECT Myocardial Blood Flow Quantitation to Guide Management for Coronary Artery Disease

Recently, myocardial blood flow quantitation with dynamic SPECT has been validated to enhance the detection of multivessel coronary artery disease (CAD) and conclude equivocal SPECT myocardial perfusion study. This advance opened an important clinical application to utilize the tool in guiding CAD management for area where myocardial perfusion tracers for PET are unavailable or unaffordable. We present a clinical patient with ongoing recursive angina who underwent multiple nuclear stress tests for a sequence of CAD evaluation in 26 months and demonstrated that SPECT myocardial blood flow quantitation properly guided CAD management to warrant patient outcome.

SPECT myocardial blood flow quantitation toward clinical use: a comparative study with 13N-Ammonia PET myocardial blood flow quantitation

OBJECTIVES

The aim of this study was to evaluate the accuracy of myocardial blood flow (MBF) quantitation of ^99mTc-Sestamibi (MIBI) single photon emission computed tomography (SPECT) compared with ¹³N-Ammonia (NH3) position emission tomography (PET) on the same cohorts.

BACKGROUND

Recent advances of SPECT technologies have been applied to develop MBF quantitation as a promising tool to diagnose coronary artery disease (CAD) for areas where PET MBF quantitation is not available. However, whether the SPECT approach can achieve the same level of accuracy as the PET approach for clinical use still needs further investigations.

METHODS

Twelve healthy volunteers (HVT) and 16 clinical patients with CAD received both MIBI SPECT and NH3 PET flow scans. Dynamic SPECT images acquired with high temporary resolution were fully corrected for physical factors and processed to quantify K1 using the standard compartmental modeling. Human MIBI tracer extraction fraction (EF) was determined by comparing MIBI K1 and NH3 flow on the HVT group and then used to convert flow values from K1 for all subjects. MIBI and NH3 flow values were systematically compared to validate the SPECT approach.

RESULTS

The human MIBI EF was determined as [1.0-0.816*exp(-0.267/MBF)]. Global and regional MBF and myocardial flow reserve (MFR) of MIBI SPECT and NH3 PET were highly correlated for all subjects (global R²: MBF = 0.92, MFR = 0.78; regional R²: MBF ≥ 0.88, MFR ≥ 0.71). No significant differences for rest flow, stress flow, and MFR between these two approaches were observed (All p ≥ 0.088). Bland-Altman plots overall revealed small bias between MIBI SPECT and NH3 PET (global: ΔMBF = -0.03Lml/min/g, ΔMFR = 0.07; regional: ΔMBF = -0.07 - 0.06 , ΔMFR = -0.02 - 0.22).

CONCLUSIONS

Quantitation with SPECT technologies can be accurate to measure myocardial blood flow as PET quantitation while comprehensive imaging factors of SPECT to derive the variability between these two approaches were fully addressed and corrected.