Reliable quantification of myocardial sympathetic innervation and regional denervation using [11C]meta-hydroxyephedrine PET.

Kai Yi WuJason G E ZeltTong WangVincent DinculescuRobert MinerCatherine LapierreNicole KapsAaryn LavalleeJennifer M RenaudJames ThackerayLisa M MielniczukShin-Yee ChenIan G BurwashJean N DaSilvaRob S B BeanlandsRobert A deKemp

Published in: European journal of nuclear medicine and molecular imaging (2019)

Quantitative measures of global and regional SNS innervation were most repeatable using the simple RI method of analysis compared with the more complex VT. Observer variability was significantly lower than the test-retest repeatability using a highly automated analysis program. These results support the use of the simple RI method for reliable analysis of HED PET images in clinical research studies for future evaluation of new therapies and for risk stratification in patients with heart failure.

Keyphrases

deep learning
computed tomography
positron emission tomography
machine learning
high resolution
high throughput
quality improvement
pet imaging
mass spectrometry
atrial fibrillation
case control