Improving detection accuracy of perfusion defect in standard dose SPECT-myocardial perfusion imaging by deep-learning denoising

2021 ◽  
Author(s):  
Junchi Liu ◽  
Yongyi Yang ◽  
Miles N. Wernick ◽  
P. Hendrik Pretorius ◽  
Piotr J. Slomka ◽  
...  
Keyword(s):  
Deep Learning ◽  
Myocardial Perfusion Imaging ◽  
Myocardial Perfusion ◽  
Perfusion Imaging ◽  
Perfusion Defect ◽  
Standard Dose ◽  
Detection Accuracy ◽  
Spect Myocardial Perfusion Imaging ◽  
Spect Myocardial Perfusion

scholarly journals Deep Learning Analysis of Polar Maps from SPECT Myocardial Perfusion Imaging for Prediction of Coronary Artery Disease

2021 ◽  
Author(s):  
Narges Zahiri ◽  
Rhona Asgari ◽  
Seid-Kazem Razavi-Ratki ◽  
Ali-Asghar parach
Keyword(s):  
Deep Learning ◽  
Myocardial Perfusion Imaging ◽  
Myocardial Perfusion ◽  
Perfusion Imaging ◽  
Learning Model ◽  
Emission Computed Tomography ◽  
Spect Myocardial Perfusion Imaging ◽  
Spect Myocardial Perfusion ◽  
Deep Learning Model ◽  
Polar Map

Abstract Purpose: This study aimed to investigate the diagnostic accuracy of deep convolutional neural networks for classifying the polar map images in Single-photon emission computed tomography (SPECT) myocardial perfusion imaging (MPI) by considering the physician’s diagnosis as reference.Methods: 3318 images of stress and rest polar maps related to patients (67% women and 33% men) who underwent 99mTc-sestamibi MPI were collected. The images were manually labeled with normal and abnormal labels according to the doctor’s diagnosis reports. The proposed deep learning model was trained using stress and rest polar maps and evaluated for prediction of obstructive disease in a stratified 5-fold cross-validation procedure.Results: The mean values of accuracy, sensitivity, accuracy, specificity, f1 score, and the area under the roc curve were 0.7562, 0.7856, 0.5748, 0.7434, 0.6646, and, 0.8450, respectively over 5 folds using both stress and rest scans. The inclusion of rest perfusion maps significantly improved AUC of the deep learning model (AUC: 0.845; 95% CI: 0.832-0.857), compared with using stress polar maps only (AUC: 0.827; 95% CI: 0.814-0.840); P < 0.05.Conclusion: The results of the present work reveal the possible applications of deep learning for polar map images classification in SPECT MPI.

scholarly journals DO CHANGES IN PERFUSION DEFECT SIZE ON SERIAL SPECT MYOCARDIAL PERFUSION IMAGING TRACK OUTCOME?

2015 ◽  
Vol 65 (10) ◽  
pp. A1245
Author(s):  
Stephanie El Hajj ◽  
Ayman Farag ◽  
Steven Bleich ◽  
Padma Priya Manapragada ◽  
Ami Iskandrian ◽  
...  
Keyword(s):  
Myocardial Perfusion Imaging ◽  
Myocardial Perfusion ◽  
Perfusion Imaging ◽  
Perfusion Defect ◽  
Defect Size ◽  
Spect Myocardial Perfusion Imaging ◽  
Perfusion Defect Size ◽  
Spect Myocardial Perfusion

scholarly journals Gated SPECT Myocardial Perfusion Imaging in female patients - an observational study

2021 ◽  
Vol 22 (2) ◽  
pp. 103-107
Author(s):  
Azmal K Sarker ◽  
Afroza Naznin ◽  
Nasreen Sultana ◽  
Faria Nasreen ◽  
Lutfun Nisa
Keyword(s):  
Myocardial Perfusion Imaging ◽  
Myocardial Perfusion ◽  
Perfusion Imaging ◽  
Perfusion Defect ◽  
Gated Spect ◽  
Nyha Class ◽  
Left Ventricular Ejection ◽  
Spect Myocardial Perfusion Imaging ◽  
Female Patients ◽  
Spect Myocardial Perfusion

Background: It had been a general observation that the number of female patients undergoing SPECT myocardial perfusion imaging (MPI) is lower per week.This study was done to find out their proportionand association among their symptoms, history of (h/o) acute coronary syndrome (ACS) and corresponding myocardial perfusion. Patients and methods:This cross sectional retrospective study was conducted in 2017 on a group of patients who underwent gated SPECT myocardial perfusion imaging (GSMPI)from February 2005 to October 2016. Clinical record files and image files of all female patients were selected for analysis. The symptoms were retrospectively categorized according to New York Heart Association (NYHA) class while the patients were categorized according to their h/o ACS and myocardial perfusion in each patient was categorized as normal or perfusion defect. Data were analyzed using SPSS v.20. Results:Among 1347 patients undergoing GSMPI 196 (14.6%) were female with a 4.4-fold increase in number over 12 years. Mean age was 51.8 ± 10.1 (27-80). Symptom categorization assigned 84 patients in NYHA class I and 112 in NYHA class II. There was no h/o ACS in 143 while 53 had a previous h/o hospitalization due to ACS. GSMPI revealed normal perfusion in 156 patients with mean left ventricular ejection fraction (LVEF) of 66.6 ± 13.6. Perfusion defect was present in 40 patients with mean LVEF of 45.5 ± 15.8. NYHA class of symptoms was not associated with perfusion status (p > 0.05).H/o ACS was found to have association both with NYHA class of symptoms and with perfusions status (p < 0.05). Conclusions: Proportion of female patients undergoing GSMPI is low. Presence of perfusion defect in female was not associated with NYHA class of symptom. Bangladesh J. Nuclear Med. 22(2): 103-107, Jul 2019

Analysis on SPECT myocardial perfusion imaging with a tool derived from dynamic programming to deep learning

Optik ◽  
2021 ◽  
Vol 240 ◽  
pp. 166842
Author(s):  
Haixing Wen ◽  
Qiuyue Wei ◽  
Jin-Long Huang ◽  
Shih-Chuan Tsai ◽  
Chi-Yen Wang ◽  
...  
Keyword(s):  
Dynamic Programming ◽  
Deep Learning ◽  
Myocardial Perfusion Imaging ◽  
Myocardial Perfusion ◽  
Perfusion Imaging ◽  
Spect Myocardial Perfusion Imaging ◽  
Spect Myocardial Perfusion
Sign in / Sign up

Export Citation Format

Share Document