Revisiting Perfusion in the Forgotten Right Ventricle: Artificial Intelligence-Enhanced Quantification on PET/CT.

Area of Science:

• Cardiovascular Imaging
• Artificial Intelligence in Medicine
• Nuclear Cardiology

Background:

• Right ventricular (RV) radiotracer uptake on perfusion imaging is a known cardiovascular risk marker.
• Quantifying RV uptake is difficult due to its thin structure and variable intensity.

Purpose of the Study:

• To develop and validate an AI-enhanced method for quantifying RV activity from CT attenuation correction (CTAC) images.
• To assess the prognostic significance of AI-derived RV activity measures for cardiovascular events.

Main Methods:

• Deep learning was used to segment the RV and left ventricular myocardium from CTAC images in 25,444 patients undergoing PET myocardial perfusion imaging.
• RV activity measures were quantified on coregistered PET images.
• Associations between RV activity and the incidence of death or myocardial infarction (MI) were evaluated.

Main Results:

• Higher maximum RV rest activity was significantly associated with an increased risk of death or MI (e.g., HR 1.17 per SD for 13N-ammonia).
• These associations remained significant after adjusting for clinical factors, perfusion, function, and myocardial flow reserve.
• The AI method successfully extracted RV activity from hybrid PET/CT imaging.

Conclusions:

• Deep learning provides a robust method for quantifying RV activity from PET/CT myocardial perfusion imaging.
• AI-derived RV activity measures offer complementary prognostic information for cardiovascular risk assessment.
• These findings highlight the potential of AI in improving cardiovascular risk stratification.