% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@ARTICLE{Full:304591,
author = {P. Full$^*$ and R. T. Schirrmeister and M. Hein and M. F.
Russe and M. Reisert and C. Ammann and K.-H. Greiser$^*$ and
T. Niendorf and T. Pischon and J. Schulz-Menger and K.
Maier-Hein$^*$ and F. Bamberg and S. Rospleszcz and C. L.
Schlett and C. Schuppert},
title = {{C}ardiac {M}agnetic {R}esonance {I}maging in the {G}erman
{N}ational {C}ohort ({NAKO}): {A}utomated {S}egmentation of
{S}hort-{A}xis {C}ine {I}mages and {P}ost-{P}rocessing
{Q}uality {C}ontrol.},
journal = {Journal of cardiovascular magnetic resonance},
volume = {nn},
issn = {1097-6647},
address = {[Amsterdam]},
publisher = {Elsevier},
reportid = {DKFZ-2025-01914},
pages = {nn},
year = {2025},
note = {#EA:E230# / epub},
abstract = {The prospective, multicenter German National Cohort (NAKO)
provides a unique dataset of cardiac magnetic resonance
(CMR) cine images. Effective processing of these images
requires a robust segmentation and quality control
pipeline.A deep learning model for semantic segmentation,
based on the nnU-Net architecture, was applied to full-cycle
short-axis cine images from 29,908 baseline participants.
The primary objective was to determine data on structure and
function for both ventricles (LV, RV), including
end-diastolic volumes (EDV), end-systolic volumes (ESV), and
LV myocardial mass. Quality control measures included a
visual assessment of outliers in morphofunctional
parameters, inter- and intra-ventricular phase differences,
and time-volume curves (TVC). These were adjudicated using a
five-point rating scale, ranging from five (excellent) to
one (non-diagnostic), with ratings of three or lower subject
to exclusion. The predictive value of outlier criteria for
inclusion and exclusion was evaluated using receiver
operating characteristics analysis.The segmentation model
generated complete data for 29,609 participants (incomplete
in $1.0\%),$ of which 5,082 cases $(17.0\%)$ underwent
visual assessment. Quality assurance yielded a sample of
26,899 $(90.8\%)$ participants with excellent or good
quality, excluding 1,875 participants due to image quality
issues and 835 participants due to segmentation quality
issues. TVC was the strongest single discriminator between
included and excluded participants (AUC: 0.684). Of the
two-category combinations, the pairing of TVC and phases
provided the greatest improvement over TVC alone (AUC
difference: 0.044; p<0.001). The best performance was
observed when all three categories were combined (AUC:
0.748). By extending the quality-controlled sample to
include mid-level 'acceptable' quality ratings, a total of
28,413 $(96.0\%)$ participants could be included.The
implemented pipeline facilitated the automated segmentation
of an extensive CMR dataset, integrating quality control
measures. This methodology ensures that ensuing quantitative
analyses are conducted with a diminished risk of bias.},
keywords = {Artificial intelligence (Other) / Cardiac MR imaging
(Other) / German National Cohort (Other) / Population
imaging (Other) / Quality control (Other)},
cin = {E230 / C020},
ddc = {610},
cid = {I:(DE-He78)E230-20160331 / I:(DE-He78)C020-20160331},
pnm = {315 - Bildgebung und Radioonkologie (POF4-315)},
pid = {G:(DE-HGF)POF4-315},
typ = {PUB:(DE-HGF)16},
pubmed = {pmid:40946969},
doi = {10.1016/j.jocmr.2025.101958},
url = {https://inrepo02.dkfz.de/record/304591},
}