Combining dynamic and ECG-gated ⁸²Rb-PET for practical implementation in the clinic

Area of Science:

• Cardiovascular Imaging
• Nuclear Cardiology
• Medical Physics

Background:

• List-mode positron emission tomography (PET) is often impractical for cardiac clinics.
• Current clinical practice relies on separate dynamic and ECG-gated acquisitions for stenosis detection, but gated studies are limited due to dose, time, and cost constraints.
• Physicians often lack comprehensive data for accurate diagnosis and treatment planning.

Purpose of the Study:

• To develop and validate a novel curve-fitting algorithm, termed incomplete data (ICD), for accurate coronary flow reserve (CFR) calculation.
• To enable CFR estimation from a combined dynamic-ECG protocol comparable in length to typical gated scans.
• To overcome the limitations of current gated PET studies in cardiac clinics.

Main Methods:

• Retrospective dynamic PET studies were used to simulate shortened dynamic acquisitions.
• The accuracy of the ICD algorithm and a nominal method was compared in extrapolating arterial input functions (AIFs).
• The accuracy of ICD and ICD with a-posteriori knowledge (ICD-AP) was evaluated in predicting CFRs against complete-data predictions.

Main Results:

• ICD-predicted arterial input functions were more accurate than the nominal method in 11 out of 12 studies.
• Coronary flow reserve (CFR) predictions by ICD and ICD-AP showed high correlation with complete-data predictions (PICD=0.94, PICD-AP=0.91).
• Average errors for ICD and ICD-AP CFR predictions were low (eICD=2.82%, eICD-AP=2.79%).

Conclusions:

• Both ICD and ICD-AP algorithms provide CFR values with sufficient accuracy for clinical use in cardiac settings.
• The developed method allows physicians to differentiate between single-vessel and triple-vessel disease.
• This facilitates informed treatment decision-making by providing comprehensive diagnostic information from a single, time-efficient scan.