Meng Chen1,2, Ximing Wang1,2, Guangyu Hao1,2, Xujie Cheng3, Chune Ma4, Ning Guo4, Su Hu1,2, Qing Tao1,2, Feirong Yao1,2, Chunhong Hu1,2. 1. Department of Radiology, The First Affiliated Hospital of Soochow University, NO.899 Pinghai Road, Gusu District, Suzhou, Jiangsu, China. 2. Institute of Medical Imaging, Soochow University, Suzhou, Jiangsu, China. 3. Department of Cardiology, The First Affiliated Hospital of Soochow University, Suzhou, Jiangsu, China. 4. ShuKun (BeiJing) Technology Co., Ltd., Jinhui Bd, Qiyang Rd, Beijing, China.
Abstract
OBJECTIVE: To investigate the diagnostic performance of deep learning (DL)-based vascular extraction and stenosis detection technology in assessing coronary artery disease (CAD). METHODS: The diagnostic performance of DL technology was evaluated by retrospective analysis of coronary computed tomography angiography in 124 suspected CAD patients, using invasive coronary angiography as reference standard. Lumen diameter stenosis ≥50% was considered obstructive, and the diagnostic performances were evaluated at per-patient, per-vessel and per-segment levels. The diagnostic performances between DL model and reader model were compared by the areas under the receiver operating characteristics curves (AUCs). RESULTS: In patient-based analysis, AUC of 0.78 was obtained by DL model to detect obstructive CAD [sensitivity of 94%, specificity of 63%, positive predictive value of 94%, and negative predictive value of 59%], While AUC by reader model was 0.74 (sensitivity of 97%, specificity of 50%, positive predictive value of 93%, negative predictive value of 73%). In vessel-based analysis, the AUCs of DL model and reader model were 0.87 and 0.89 respectively. In segment-based analysis, the AUCs of 0.84 and 0.89 were obtained by DL model and reader model respectively. It took 0.47 min to analyze all segments per patient by DL model, which is significantly less than reader model (29.65 min) (p < 0.001). CONCLUSION: The DL technology can accurately and effectively identify obstructive CAD, with less time-consuming, and it could be a reliable diagnostic tool to detect CAD. ADVANCES IN KNOWLEDGE: The DL technology has valuable prospect with the diagnostic ability to detect CAD.
OBJECTIVE: To investigate the diagnostic performance of deep learning (DL)-based vascular extraction and stenosis detection technology in assessing coronary artery disease (CAD). METHODS: The diagnostic performance of DL technology was evaluated by retrospective analysis of coronary computed tomography angiography in 124 suspected CAD patients, using invasive coronary angiography as reference standard. Lumen diameter stenosis ≥50% was considered obstructive, and the diagnostic performances were evaluated at per-patient, per-vessel and per-segment levels. The diagnostic performances between DL model and reader model were compared by the areas under the receiver operating characteristics curves (AUCs). RESULTS: In patient-based analysis, AUC of 0.78 was obtained by DL model to detect obstructive CAD [sensitivity of 94%, specificity of 63%, positive predictive value of 94%, and negative predictive value of 59%], While AUC by reader model was 0.74 (sensitivity of 97%, specificity of 50%, positive predictive value of 93%, negative predictive value of 73%). In vessel-based analysis, the AUCs of DL model and reader model were 0.87 and 0.89 respectively. In segment-based analysis, the AUCs of 0.84 and 0.89 were obtained by DL model and reader model respectively. It took 0.47 min to analyze all segments per patient by DL model, which is significantly less than reader model (29.65 min) (p < 0.001). CONCLUSION: The DL technology can accurately and effectively identify obstructive CAD, with less time-consuming, and it could be a reliable diagnostic tool to detect CAD. ADVANCES IN KNOWLEDGE: The DL technology has valuable prospect with the diagnostic ability to detect CAD.
Authors: Maria D'Souza; Laura Sarkisian; Lotte Saaby; Tina S Poulsen; Oke Gerke; Torben B Larsen; Axel C P Diederichsen; Nikolaj Jangaard; Søren Z Diederichsen; Susanne Hosbond; Jens Hove; Kristian Thygesen; Hans Mickley Journal: Am J Med Date: 2015-03-27 Impact factor: 4.965
Authors: Stewart M Benton; Christian Tesche; Carlo N De Cecco; Taylor M Duguay; U Joseph Schoepf; Richard R Bayer Journal: J Thorac Imaging Date: 2018-03 Impact factor: 3.000
Authors: David E Newby; Philip D Adamson; Colin Berry; Nicholas A Boon; Marc R Dweck; Marcus Flather; John Forbes; Amanda Hunter; Stephanie Lewis; Scott MacLean; Nicholas L Mills; John Norrie; Giles Roditi; Anoop S V Shah; Adam D Timmis; Edwin J R van Beek; Michelle C Williams Journal: N Engl J Med Date: 2018-08-25 Impact factor: 91.245
Authors: W Bob Meijboom; Matthijs F L Meijs; Joanne D Schuijf; Maarten J Cramer; Nico R Mollet; Carlos A G van Mieghem; Koen Nieman; Jacob M van Werkhoven; Gabija Pundziute; Annick C Weustink; Alexander M de Vos; Francesca Pugliese; Benno Rensing; J Wouter Jukema; Jeroen J Bax; Mathias Prokop; Pieter A Doevendans; Myriam G M Hunink; Gabriel P Krestin; Pim J de Feyter Journal: J Am Coll Cardiol Date: 2008-12-16 Impact factor: 24.094
Authors: Jan Menke; Christina Unterberg-Buchwald; Wieland Staab; Jan Martin Sohns; Ali Seif Amir Hosseini; Alexander Schwarz Journal: Am Heart J Date: 2012-11-26 Impact factor: 4.749
Authors: Carlos Cano-Espinosa; Germán González; George R Washko; Miguel Cazorla; Raúl San José Estépar Journal: Proc SPIE Int Soc Opt Eng Date: 2018-03-02
Authors: Bjarne L Nørgaard; Jonathon Leipsic; Sara Gaur; Sujith Seneviratne; Brian S Ko; Hiroshi Ito; Jesper M Jensen; Laura Mauri; Bernard De Bruyne; Hiram Bezerra; Kazuhiro Osawa; Mohamed Marwan; Christoph Naber; Andrejs Erglis; Seung-Jung Park; Evald H Christiansen; Anne Kaltoft; Jens F Lassen; Hans Erik Bøtker; Stephan Achenbach Journal: J Am Coll Cardiol Date: 2014-01-30 Impact factor: 24.094