Page 58 - Handbook of Deep Learning in Biomedical Engineering Techniques and Applications

Article type Journal Conference Conference Journal Conference

References [53] [109] [110] [111] [112]

machine China, York by volumes collected three scans and Philips The scanned coronary
radiographs radiography provided MRI were Medical, patients, into the train 5 LV, segmentation, the testing at exams Best, CTA from have to

chest Guangzhou, study set cardiac taken scans Philips iCT, 60 of divided to used the of for CT Medical, consecutively annotated acquired suspected
set digital data was randomly were LV scans cardiac Philips 250 used were or disease
data a Hospital, the for available containing angiography Brilliance Netherlands, localization train to 5 iCT, of which known artery

on posterioreanterior with Nanfang used patients CT Philips The were scans 50 for used remaining obtained Netherlands, were comprehensively sets,
Remark 646 acquired at were publicly A University 33 for Cardiac at Best, which sets: CNNs were the method Clinically Brilliance patients 110 data patients coronary

imaging.dcontinued Modality CNN Chest CNN radiography MRI angiography CT angiography CT (CTA) CTA

medical of Type Multiscale CNN Deep AlexNet CNN Paired (CovPairs) CNN

to in in coronary CT centerlines
application bone study conventional automatic ventricle left automatic ventricle left automatic cardiac in automatic vessel

CNN to single in for the of Images for the of angiography for scoring for blood

of designed suppression X-ray designed localization MRI designed segmentation CT designed calcium angiography designed of
Literature Applications Model chest Model Cardiac System cardiac System artery System extraction

2.2 Year 2017 2015 2016
Table Body part Cardiac