scholarly journals Machine learning predicts per-vessel early coronary revascularization after fast myocardial perfusion SPECT: results from multicentre REFINE SPECT registry

2019 ◽  
Vol 21 (5) ◽  
pp. 549-559 ◽  
Author(s):  
Lien-Hsin Hu ◽  
Julian Betancur ◽  
Tali Sharir ◽  
Andrew J Einstein ◽  
Sabahat Bokhari ◽  
...  
Keyword(s):  
Machine Learning ◽  
Myocardial Perfusion ◽  
Single Photon ◽  
Coronary Revascularization ◽  
Stress Test ◽  
Invasive Coronary Angiography ◽  
Photon Emission ◽  
Myocardial Perfusion Spect ◽  
Patient Specific ◽  
Emission Computed Tomography

Abstract Aims To optimize per-vessel prediction of early coronary revascularization (ECR) within 90 days after fast single-photon emission computed tomography (SPECT) myocardial perfusion imaging (MPI) using machine learning (ML) and introduce a method for a patient-specific explanation of ML results in a clinical setting. Methods and results A total of 1980 patients with suspected coronary artery disease (CAD) underwent stress/rest 99mTc-sestamibi/tetrofosmin MPI with new-generation SPECT scanners were included. All patients had invasive coronary angiography within 6 months after SPECT MPI. ML utilized 18 clinical, 9 stress test, and 28 imaging variables to predict per-vessel and per-patient ECR with 10-fold cross-validation. Area under the receiver operator characteristics curve (AUC) of ML was compared with standard quantitative analysis [total perfusion deficit (TPD)] and expert interpretation. ECR was performed in 958 patients (48%). Per-vessel, the AUC of ECR prediction by ML (AUC 0.79, 95% confidence interval (CI) [0.77, 0.80]) was higher than by regional stress TPD (0.71, [0.70, 0.73]), combined-view stress TPD (AUC 0.71, 95% CI [0.69, 0.72]), or ischaemic TPD (AUC 0.72, 95% CI [0.71, 0.74]), all P < 0.001. Per-patient, the AUC of ECR prediction by ML (AUC 0.81, 95% CI [0.79, 0.83]) was higher than that of stress TPD, combined-view TPD, and ischaemic TPD, all P < 0.001. ML also outperformed nuclear cardiologists’ expert interpretation of MPI for the prediction of early revascularization performance. A method to explain ML prediction for an individual patient was also developed. Conclusion In patients with suspected CAD, the prediction of ECR by ML outperformed automatic MPI quantitation by TPDs (per-vessel and per-patient) or nuclear cardiologists’ expert interpretation (per-patient).

scholarly journals A case report: adenosine triggered myocardial infarction during myocardial perfusion stress test imaging in a diabetic patient

2021 ◽  
Vol 5 (4) ◽  
Author(s):  
Maarten de Mulder ◽  
Menno van Gameren ◽  
Eric A van Asperen ◽  
Martijn Meuwissen
Keyword(s):  
Myocardial Infarction ◽  
Myocardial Perfusion ◽  
Single Photon ◽  
Thrombus Formation ◽  
Stress Test ◽  
Photon Emission ◽  
Coronary Perfusion Pressure ◽  
Emission Computed Tomography ◽  
Coronary Perfusion ◽  
Starting Point

Abstract Background Myocardial perfusion imaging (MPI) using single-photon emission computed tomography (SPECT) can in general be used safely in daily clinical practice. However, under the right circumstances, it can lead to serious complications. Case summary A 68-year-old female patient with diabetes and a history of inferior ST-elevation myocardial infarction 8 years earlier, visited our outpatient clinic with atypical chest discomfort. In order to assess whether this is due to myocardial ischaemia, MPI-SPECT was ordered. As it was suspected she would not achieve sufficient exercise levels, pharmacologic stress using adenosine was arranged. During the scan, she developed acute myocardial infarction. Subsequent urgent coronary angiography demonstrated a subtotal stenosis in the proximal left anterior descending coronary artery which was successfully stented. She was still free from angina 4 months later. Discussion The combination of a reduced systemic and coronary perfusion pressure in the presence of an exhausted coronary autoregulation, may be a starting point for local geometrical changes that initiate the classic cascade of thrombus formation and acute occlusion of coronary arteries during MPI-SPECT. This illustrates the need for continuous patient and electrocardiogram monitoring.

scholarly journals Machine Learning Algorithms to Distinguish Myocardial Perfusion SPECT Polar Maps

2021 ◽  
Vol 8 ◽  
Author(s):  
Erito Marques de Souza Filho ◽  
Fernando de Amorim Fernandes ◽  
Christiane Wiefels ◽  
Lucas Nunes Dalbonio de Carvalho ◽  
Tadeu Francisco dos Santos ◽  
...  
Keyword(s):  
Machine Learning ◽  
Myocardial Perfusion ◽  
Single Photon ◽  
Photon Emission ◽  
Image Features ◽  
Machine Learning Algorithms ◽  
University Hospital ◽  
Emission Computed Tomography ◽  
Operating Characteristics ◽  
Adaptive Boosting

Myocardial perfusion imaging (MPI) plays an important role in patients with suspected and documented coronary artery disease (CAD). Machine Learning (ML) algorithms have been developed for many medical applications with excellent performance. This study used ML algorithms to discern normal and abnormal gated Single Photon Emission Computed Tomography (SPECT) images. We analyzed one thousand and seven polar maps from a database of patients referred to a university hospital for clinically indicated MPI between January 2016 and December 2018. These studies were reported and evaluated by two different expert readers. The image features were extracted from a specific type of polar map segmentation based on horizontal and vertical slices. A senior expert reading was the comparator (gold standard). We used cross-validation to divide the dataset into training and testing subsets, using data augmentation in the training set, and evaluated 04 ML models. All models had accuracy >90% and area under the receiver operating characteristics curve (AUC) >0.80 except for Adaptive Boosting (AUC = 0.77), while all precision and sensitivity obtained were >96 and 92%, respectively. Random Forest had the best performance (AUC: 0.853; accuracy: 0,938; precision: 0.968; sensitivity: 0.963). ML algorithms performed very well in image classification. These models were capable of distinguishing polar maps remarkably into normal and abnormal.

scholarly journals Quantitative analysis of gated myocardial perfusion SPECT

2018 ◽  
pp. 220-227
Author(s):  
Victor Ploskikh ◽  
Elena Kotina
Keyword(s):  
Quantitative Analysis ◽  
Myocardial Perfusion ◽  
Data Processing ◽  
Single Photon ◽  
Photon Emission ◽  
Myocardial Perfusion Spect ◽  
Mathematical Representation ◽  
Emission Computed Tomography ◽  
Gated Myocardial Perfusion Spect ◽  
Perfusion Spect

The paper considers the problem of gated myocardial perfusion single photon emission computed tomography (SPECT) data processing. An approach to the quantitative analysis of gated myocardial perfusion SPECT data used in software developed in the SPbSU is presented. The article presents and formalizes the complete data processing workflow. All the main tasks of the data processing are considered. Mathematical representation of problem domain objects is presented. A detailed algorithm of the data processing is given. The algorithmis implemented as component of the data processing software suite.

scholarly journals Patients undergoing invasive coronary angiography after a positive single-photon emission computed tomography or a positive stress cardiac magnetic resonance - What to expect at the cath lab

2021 ◽  
Vol 22 (Supplement_2) ◽  
Author(s):  
C Silva ◽  
M Goncalves ◽  
P Lopes ◽  
A Ventosa ◽  
J Calqueiro ◽  
...  
Keyword(s):  
Single Photon ◽  
Stress Test ◽  
Invasive Coronary Angiography ◽  
Photon Emission ◽  
Emission Computed Tomography ◽  
Predictive Values ◽  
Tertiary Center ◽  
Single Photon Emission Computed ◽  
Stress Cmr ◽  
Photon Emission Computed Tomography

Abstract Funding Acknowledgements Type of funding sources: None. Background  Randomized controlled trials comparing stress cardiac magnetic resonance (CMR) and single-photon emission computed tomography (SPECT) suggest similar diagnostic accuracy for detecting obstructive coronary artery disease (CAD). However, there are few data on whether or not this remains true in routine clinical practice. The aim of this study was to assess the clinical and angiographic characteristics of patients undergoing invasive coronary angiography (ICA) after a positive stress CMR or positive SPECT, and to compare their positive predictive value with published results from the CE-MARC trial. Methods In this retrospective tertiary-center analysis, we included 429 patients (mean age 67 ± 10 years, 28% women, 42% diabetic) undergoing ICA between January 2016 and December 2020, after a positive stress CMR or positive SPECT. Regarding stress test, an adenosine protocol was performed in all stress CMR and in 76.4% (n = 272) of stress SPECT.  Stress test results, including ischemia location and severity, were classified as reported by their primary readers. Patients with missing data on key variables, and those in whom microvascular disease was considered likely in the original stress test report were excluded. Obstructive CAD was defined as any coronary artery stenosis ≥ 50% in a vessel compatible with the ischemic territory on stress testing. Results Out of the total 429 patients, 356 (83%) were referred after a positive SPECT, and 73 (17%) after a positive stress CMR. Patients did not differ regarding age, cardiovascular risk factors, previous revascularization or left ventricular dysfunction, but patients with SPECT were more frequently male (p = 0.046). Overall, 320 patients (75%) had obstructive CAD on ICA. The prevalence of obstructive CAD was similar in patients with positive SPECT vs. positive stress CMR (76.1% vs. 80.8%, respectively, p = 0.385). There were also no significant differences in the prevalence of left main or 3-vessel disease (9.0% vs. 9.6%, p = 0.871, and 19.7% vs. 23.3% p = 0.483, respectively). Revascularization was performed or planned in 59.3% of patients in the SPECT group, and 52.1% of those in the stress CMR group (p = 0.255). The positive predictive values of both techniques were similar to those reported in the CE-MARC trial (Figure), and would increase to 88.1% and 89.4% for SPECT and stress CMR, respectively, if patients reported as having only mild ischemia were excluded. Conclusion In this tertiary center analysis, stress CMR and SPECT showed similar positive predictive values, comparable to those reported in the CE-MARC trial.

scholarly journals Challenges of gated myocardial perfusion SPECT processing

2021 ◽  
pp. 171-177
Author(s):  
Victor Ploskikh ◽  
Elena Kotina
Keyword(s):  
Myocardial Perfusion ◽  
Single Photon ◽  
Photon Emission ◽  
Myocardial Perfusion Spect ◽  
Mathematical Representation ◽  
Emission Computed Tomography ◽  
Gated Myocardial Perfusion Spect ◽  
Myocardial Perfusion Single Photon ◽  
Perfusion Spect ◽  
Photon Emission Computed Tomography

The paper provides an in-depth look at gated myocardial perfusion single photon emission computed tomography (SPECT) data processing. Attention paid to several unmentioned subjects of the quantitative analysis of gated myocardial perfusion SPECT data. The article considers several options in the construction process of the ellipsoid coordinate system of the left ventricle (LV). Mathematical representation of polar maps is given. Formulas of the regional parameters calculation are proposed. Issues of phase analysis are explored.

scholarly journals Radiomics analysis of myocardial perfusion SPECT images in patients with cardiomyopathy and heart failure

2021 ◽  
Vol 22 (Supplement_3) ◽  
Author(s):  
K Okuda ◽  
K Nakajima ◽  
H Saito ◽  
S Yamashita ◽  
M Hashimoto ◽  
...  
Keyword(s):  
Heart Failure ◽  
Myocardial Perfusion ◽  
Single Photon ◽  
Photon Emission ◽  
Myocardial Perfusion Spect ◽  
Feature Reduction ◽  
Support Vector ◽  
Emission Computed Tomography ◽  
Operating Characteristics ◽  
Perfusion Spect

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): JSPS KAKENHI Grants Background Although myocardial perfusion heterogeneity due to focally damaged cardiomyocytes is observed in single−photon emission computed tomography (SPECT) imaging, a current perfusion defect scoring system does not allow us to provide sufficient diagnostic information for heterogeneity. Purpose The aim of this study was to perform radiomics analysis of myocardial perfusion SPECT (MPS) images to investigate the potential to detect myocardial perfusion heterogeneity. Methods Patients with hypertrophic cardiomyopathy (n = 3), heart failure (n = 9), and with a low likelihood of coronary artery disease (n =15) (Figure 1), who underwent a rest 99mTc-MIBI myocardial perfusion SPECT, were assessed using a LIFEx software. Four shape−based features, 6 histogram−based features, and 32 textural features were computed. The relevant features for the classification of the patients were selected using the Boruta algorithm, and hierarchical clustering of the selected features using the Spearman correlation coefficient was also performed for the feature reduction. The receiver operating characteristics (ROC) analysis was performed by the support vector machine to calculate the area under the ROC curve (AUC) for the selected features. Results Of 40 features, 17 were selected by the classification analysis, and these features were classified into 7 classes by the correlation analysis (Figure 2). The ROC AUCs for 7 features extracted from each class were 0.99, 0.97, 0.96, 0.92, 0.90, 0.86, and 0.83 for the contrast of NDGLDM, the entropy of histogram, ZLNU of GLZLM, the energy of GLCM, the energy of histogram, SZLGE of GLZLM, and the correlation of GLCM, respectively, as compared to 0.39 for a summed rest score. Conclusions Radiomics analysis successfully determined the myocardial perfusion heterogeneity in patients with cardiomyopathy and heart failure. It might be promising for the evaluation of myocardial damages that cannot be analyzed by the conventional scoring method.

Abstract 15252: Segmental Myocardial Perfusion Improves after Repeated Intramyocardial Cell Injection

Circulation ◽  
2014 ◽  
Vol 130 (suppl_2) ◽  
Author(s):  
Imke Mann ◽  
Sander F Rodrigo ◽  
Jan van Ramshorst ◽  
Saskia L Beeres ◽  
Jaap J Zwaginga ◽  
...  
Keyword(s):  
Myocardial Perfusion ◽  
Single Photon ◽  
Photon Emission ◽  
Emission Computed Tomography ◽  
Bone Marrow Mononuclear Cell ◽  
Repeated Injection ◽  
Refractory Angina ◽  
Rest Perfusion ◽  
Refractory Angina Pectoris ◽  
First Time

Introduction: We previously demonstrated that intramyocardial bone marrow mononuclear cell (MNC) injection improves segmental myocardial perfusion. This study was designed to evaluate in patients with recurrent refractory angina the effect of repeated injection on segmental myocardial perfusion. Methods: Twenty-one patients with recurrent refractory angina pectoris, who received 100x10^6 autologous MNC intramyocardially using the NOGA-system for a second time, were enrolled. Single-photon emission computed tomography was performed at baseline and 3 months after both injection procedures. The myocardium was divided into 17 segments and in both stress and rest images, segmental tracer activity was categorized on a 4-point scale. (1=>75%; 2=50%-74%; 3=25%-49%; 4=<25%) Segments demonstrating increased perfusion of at least 1 point in stress or rest perfusion were categorized as improved. Results: The second injection procedure was 4.6 ± 2.5 years after the first. In total, 139 segments were injected for the first time during either of the procedures, of which 80(58%) segments improved. Repeated injection in the same segment was performed in 45 segments. Of these segments, 18(40%) improved, less than after a first injection (P=0.030). Repeatedly injected segments can be subdivided in 29 previously responding segments (improved after the first injection) and 16 previously non-responding segments (not improved after the first injection). Of the responding segments, 13(45%) segments improved after repeated injection and of the non-responding segments, 5(31%) segments improved. This difference was not significant (P=.476). Conclusions: Segmental myocardial perfusion can improve after repeated intramyocardial MNC injection independently of the effect of the first injection, but the first injection is more effective.

Sign in / Sign up

Export Citation Format

Share Document