Page 210 - Artificial Intelligence for Computational Modeling of the Heart

6

Additional clinical applications

d
c
a,b
d
Felix Meister , Helene Houle , Cosmin Nita , Andrei Puiu ,
d
Lucian Mihai Itu , Saikiran Rapaka a
b
a Siemens Healthineers, Princeton, NJ, United States. Friedrich-Alexander
c
University Erlangen-Nuremberg, Erlangen, Germany. Siemens Healthineers,
d
Ultrasound Division, Mountain View, CA, United States. Siemens SRL, Image
Fusion and Analytics, Brasov, Romania
6.1 Cardiac resynchronization therapy
6.1.1 Introduction
Heart failure (HF) is a cardiovascular disease (CVD) charac-
terized by a reduced cardiac output due to a systolic or diastolic
dysfunction [389,390]. Consequently, the heart cannot meet the
body demand for oxygen and nutrients. Over the last decade, an
increasing prevalence of HF has been observed, which has been
linked to a growth of the elderly population and higher survival
rates to acute myocardial infarction. To treat the failing heart, car-
diac resynchronization therapy (CRT) proved to be efficient for
patients that suffer from a dyssynchonous cardiac contraction
caused by dilated cardiomyopathy (DCM). A set of electrodes is
implanted to counteract arrhythmias and to resynchronize the
beating cardiac chambers by emitting precisely timed electrical
impulses. Several benefits associated with successful CRT treat-
ment have been reported, including increased quality of life and
reduced total mortality [391].
Even though CRT effectiveness has been demonstrated in sev-
eral studies, it has also been observed that 30 to 50% of treated
patients do not respond to this therapy [392–395]. Several fac-
tors have been identified, for instance of clinical nature or related
to electrical properties of the diseased hearts. However, biomark-
ers that can reliably identify responders have not been identified
so far [396–400]. Other related studies suggest that new imag-
ing technologies like speckle tracking echocardiography [401–403]
and cardiac MRI with late enhancement [404–406] may help in
predicting the response prior to the intervention. In general, these
approaches aim at improving the patient selection using baseline
information, and hence neglect the effect of the procedure itself.

Artificial Intelligence for Computational Modeling of the Heart 183
https://doi.org/10.1016/B978-0-12-817594-1.00017-6
Copyright © 2020 Elsevier Inc. All rights reserved.