BibTex format
@article{Cheng:2025:10.1016/j.jocmr.2024.101133,
author = {Cheng, L-H and Sun, X and Elliot, C and Condliffe, R and Kiely, DG and Alabed, S and Swift, AJ and van, der Geest RJ and Kiely, DG and Watson, L and Armstrong, I and Billings, C and Charalampopoulos, A and Condliffe, R and Elliot, C and Hameed, A and Hamilton, N and Hurdman, J and Lawrie, A and Lewis, RA and Rajaram, S and Rothman, A and Swift, AJ and Wood, S and Roger, Thompson AA and Wild, J},
doi = {10.1016/j.jocmr.2024.101133},
journal = {Journal of Cardiovascular Magnetic Resonance},
title = {Mean pulmonary artery pressure prediction with explainable multi-view cardiovascular magnetic resonance cine series deep learning model},
url = {http://dx.doi.org/10.1016/j.jocmr.2024.101133},
volume = {27},
year = {2025}
}
RIS format (EndNote, RefMan)
TY - JOUR
AB - BackgroundPulmonary hypertension (PH) is a heterogeneous condition and regardless of etiology impacts negatively on survival. Diagnosis of PH is based on hemodynamic parameters measured invasively at right heart catheterization (RHC); however, a non-invasive alternative would be clinically valuable. Our aim was to estimate RHC parameters non-invasively from cardiac magnetic resonance (MR) data using deep learning models and to identify key contributing imaging features.MethodsWe constructed an explainable convolutional neural network (CNN) taking cardiac MR cine series from four different views as input to predict mean pulmonary artery pressure (mPAP). The model was trained and evaluated on 1646 examinations. The model’s attention weight and predictive performance associated with each frame, view, or phase were used to judge its importance. Additionally, the importance of each cardiac chamber was inferred by perturbing part of the input pixels.ResultsThe model achieved a Pearson correlation coefficient of 0.80 and R2 of 0.64 in predicting mPAP and identified the right ventricle region on short-axis view to be especially informative.ConclusionHemodynamic parameters can be estimated non-invasively with a CNN, using MR cine series from four views, revealing key contributing features at the same time.
AU - Cheng,L-H
AU - Sun,X
AU - Elliot,C
AU - Condliffe,R
AU - Kiely,DG
AU - Alabed,S
AU - Swift,AJ
AU - van,der Geest RJ
AU - Kiely,DG
AU - Watson,L
AU - Armstrong,I
AU - Billings,C
AU - Charalampopoulos,A
AU - Condliffe,R
AU - Elliot,C
AU - Hameed,A
AU - Hamilton,N
AU - Hurdman,J
AU - Lawrie,A
AU - Lewis,RA
AU - Rajaram,S
AU - Rothman,A
AU - Swift,AJ
AU - Wood,S
AU - Roger,Thompson AA
AU - Wild,J
DO - 10.1016/j.jocmr.2024.101133
PY - 2025///
SN - 1097-6647
TI - Mean pulmonary artery pressure prediction with explainable multi-view cardiovascular magnetic resonance cine series deep learning model
T2 - Journal of Cardiovascular Magnetic Resonance
UR - http://dx.doi.org/10.1016/j.jocmr.2024.101133
UR - https://doi.org/10.1016/j.jocmr.2024.101133
VL - 27
ER -