Fully automated calculation of image-derived input function in simultaneous PET/MRI in a sheep model

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Biophysics

Background:

• Dynamic positron emission tomography (PET) requires an arterial input function (AIF) for accurate quantification.
• Arterial blood sampling is invasive; image-derived input function (IDIF) offers a non-invasive alternative.
• Simultaneous PET/MRI allows high-resolution angiography for PVE and PSF correction.

Purpose of the Study:

• To develop and validate a fully automated method for obtaining the IDIF in PET/MRI.
• To compare the performance of the automated IDIF method against manual arterial blood sampling.
• To assess the accuracy of IDIF in quantifying cerebral blood flow (CBF).

Main Methods:

• Segmentation of major neck arteries using enhanced MRI angiography.
• Application of a vessel-enhancement filter based on the inertia tensor.
• Geometrical deconvolution using segmented arterial mask and measured PET point spread function (PSF).

Main Results:

• A cross-calibration factor of 0.98 ± 0.10 was achieved between IDIF and blood samples during steady state.
• The IDIF demonstrated an earlier and sharper bolus peak compared to arterial blood sampling.
• Cerebral blood flow (CBF) calculated using IDIF was 22% higher than that from arterial blood sampling.

Conclusions:

• The automated IDIF method effectively corrects for partial-volume effects (PVE) and PSF.
• Discrepancies in CBF values are attributed to differences in sampling site delay and dispersion.
• Further research will focus on human studies to replace invasive blood sampling with IDIF in PET/MRI.