Chi-Lun Ko1,2,3, Shau-Syuan Lin1, Cheng-Wen Huang1, Yu-Hui Chang1, Kuan-Yin Ko4, Mei-Fang Cheng2,3, Shan-Ying Wang5, Chung-Ming Chen1, Yen-Wen Wu6,7,8,9,10,11. 1. Department of Biomedical Engineering, National Taiwan University, Taipei, Taiwan. 2. Department of Nuclear Medicine, National Taiwan University Hospital, Taipei, Taiwan. 3. College of Medicine, National Taiwan University, Taipei, Taiwan. 4. Department of Nuclear Medicine, National Taiwan University Cancer Center, Taipei, Taiwan. 5. Department of Nuclear Medicine, Far Eastern Memorial Hospital, New Taipei City, Taiwan. 6. Department of Nuclear Medicine, National Taiwan University Hospital, Taipei, Taiwan. wuyw0502@gmail.com. 7. College of Medicine, National Taiwan University, Taipei, Taiwan. wuyw0502@gmail.com. 8. Department of Nuclear Medicine, Far Eastern Memorial Hospital, New Taipei City, Taiwan. wuyw0502@gmail.com. 9. Division of Cardiology, Cardiovascular Medical Center, Far Eastern Memorial Hospital, No. 21, Sec. 2, Nanya S. Rd., Banciao Dist., New Taipei City, 220, Taiwan. wuyw0502@gmail.com. 10. School of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan. wuyw0502@gmail.com. 11. Graduate Institute of Medicine, Yuan Ze University, Taoyuan, Taiwan. wuyw0502@gmail.com.
Abstract
PURPOSE: Deep learning (DL) models have been shown to outperform total perfusion deficit (TPD) quantification in predicting obstructive coronary artery disease (CAD) from myocardial perfusion imaging (MPI). However, previously published methods have depended on polar maps, required manual correction, and normal database. In this study, we propose a polar map-free 3D DL algorithm to predict obstructive disease. METHODS: We included 1861 subjects who underwent MPI using cadmium-zinc-telluride camera and subsequent coronary angiography. The subjects were divided into parameterization and external validation groups. We implemented a fully automatic algorithm to segment myocardium, perform registration, and apply normalization. We further flattened the image based on spherical coordinate system transformation. The proposed model consisted of a component to predict patent arteries and a component to predict disease in each vessel. The model was cross-validated in the parameterization group, and then further tested using the external validation group. The performance was assessed by area under receiver operating characteristic curves (AUCs) and compared with TPD. RESULTS: Our algorithm preprocessed all images accurately as confirmed by visual inspection. In patient-based analysis, the AUC of the proposed model was significantly higher than that for stress-TPD (0.84 vs 0.76, p < 0.01). In vessel-based analysis, the proposed model also outperformed regional stress-TPD (AUC = 0.80 vs 0.72, p < 0.01). The addition of quantitative images did not improve the performance. CONCLUSIONS: Our proposed polar map-free 3D DL algorithm to predict obstructive CAD from MPI outperformed TPD and did not require manual correction or a normal database.
PURPOSE: Deep learning (DL) models have been shown to outperform total perfusion deficit (TPD) quantification in predicting obstructive coronary artery disease (CAD) from myocardial perfusion imaging (MPI). However, previously published methods have depended on polar maps, required manual correction, and normal database. In this study, we propose a polar map-free 3D DL algorithm to predict obstructive disease. METHODS: We included 1861 subjects who underwent MPI using cadmium-zinc-telluride camera and subsequent coronary angiography. The subjects were divided into parameterization and external validation groups. We implemented a fully automatic algorithm to segment myocardium, perform registration, and apply normalization. We further flattened the image based on spherical coordinate system transformation. The proposed model consisted of a component to predict patent arteries and a component to predict disease in each vessel. The model was cross-validated in the parameterization group, and then further tested using the external validation group. The performance was assessed by area under receiver operating characteristic curves (AUCs) and compared with TPD. RESULTS: Our algorithm preprocessed all images accurately as confirmed by visual inspection. In patient-based analysis, the AUC of the proposed model was significantly higher than that for stress-TPD (0.84 vs 0.76, p < 0.01). In vessel-based analysis, the proposed model also outperformed regional stress-TPD (AUC = 0.80 vs 0.72, p < 0.01). The addition of quantitative images did not improve the performance. CONCLUSIONS: Our proposed polar map-free 3D DL algorithm to predict obstructive CAD from MPI outperformed TPD and did not require manual correction or a normal database.
Authors: Stephan D Fihn; James C Blankenship; Karen P Alexander; John A Bittl; John G Byrne; Barbara J Fletcher; Gregg C Fonarow; Richard A Lange; Glenn N Levine; Thomas M Maddox; Srihari S Naidu; E Magnus Ohman; Peter K Smith Journal: J Am Coll Cardiol Date: 2014-07-28 Impact factor: 24.094
Authors: Juhani Knuuti; William Wijns; Antti Saraste; Davide Capodanno; Emanuele Barbato; Christian Funck-Brentano; Eva Prescott; Robert F Storey; Christi Deaton; Thomas Cuisset; Stefan Agewall; Kenneth Dickstein; Thor Edvardsen; Javier Escaned; Bernard J Gersh; Pavel Svitil; Martine Gilard; David Hasdai; Robert Hatala; Felix Mahfoud; Josep Masip; Claudio Muneretto; Marco Valgimigli; Stephan Achenbach; Jeroen J Bax Journal: Eur Heart J Date: 2020-01-14 Impact factor: 29.983
Authors: Francesco Nudi; Ami E Iskandrian; Orazio Schillaci; Mariangela Peruzzi; Giacomo Frati; Giuseppe Biondi-Zoccai Journal: JACC Cardiovasc Imaging Date: 2017-03-15
Authors: Rory Hachamovitch; Alan Rozanski; Leslee J Shaw; Gregg W Stone; Louise E J Thomson; John D Friedman; Sean W Hayes; Ishac Cohen; Guido Germano; Daniel S Berman Journal: Eur Heart J Date: 2011-01-21 Impact factor: 29.983
Authors: Elsemiek M Engbers; Jorik R Timmer; Mohamed Mouden; Siert Knollema; Pieter L Jager; Jan Paul Ottervanger Journal: J Nucl Med Date: 2017-04-27 Impact factor: 10.057