Page 59 - Handbook of Deep Learning in Biomedical Engineering Techniques and Applications

Conference Conference Conference Conference Conference Journal Journal

[113] [114] [115] [116] [117] [118] [119]

patients axial training was sets Cardiac the for acquired MR left comprising of mix

75 2D study ultrasound for sets data the SATA used end-diastolic cardiac 2009 a
from of the images data pilot by was images subjects 45 MICCAI challenge used having
CT hundreds for five on MRI provided 2013 on MR using the was patients,
cardiac of used were done 34,361 testing Biobank UK 100 of set MICCAI Challenge evaluated cardiac adult done was from segmentation set data 45

included consisting which were total with for the of consisting image in Project was cine of healthy 1233 taken from conditions injury.

set each slices, Experiments views 8,533 subset used MR Atlas Segmentation study model frames from evaluation sets data ventricle Sunnybrook MRI cine cardiac brain
Data and A Cardiac The The The traumatic

TBI,
imaging;

CT US MRI MRI Net MRI MRI MRI resonance

CNN Convolutional Network fusion SuperResolution magnetic
Deep Fully (FCN) CNN Deep CNN CNN FCNN MRI,

images quality and cardiac in cardiac
given of selection ventricle cardiac timely of tomography;
automatic a for detection and ultrasound automatic MRI atlas multiatlas multiinput automatic left automatic for MRI management computed

for position slice for structure from for cardiac of for in for superresolution for the of for from CT,
designed of CT designed anatomical segmentation developed assessment designed fusion segmentation developed designed segmentation MRI designed segmentation and pathologies network;

System detection cardiac Model Model Model label Model image Model cardiac Model diagnosis neural

2017 convolutional

CNN,