Patient-dependent count-rate adaptive normalization for PET detector efficiency with delayed-window coincidence...

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Radiological Physics

Background:

• Quantitative Positron Emission Tomography (PET) imaging is crucial for clinical diagnosis in oncology and neuroimaging.
• Accurate normalization correction is essential for precise quantitative PET image reconstruction.
• Existing methods can suffer from artifacts like 'ring' artifacts due to mismatched system count-rates.

Purpose of the Study:

• To propose a new normalization calibration method for quantitative PET imaging.
• To reduce 'ring' artifacts and improve image uniformity.
• To simplify the normalization calibration procedure.

Main Methods:

• Utilizing delayed-window coincidence events from scanning phantoms or patients for normalization calibration.
• Developing a modified algorithm for mean detector efficiency estimation to create uniform crystal efficiency maps.
• Evaluating the method using both phantom and real patient datasets.

Main Results:

• The proposed method significantly reduces 'ring' artifacts caused by mismatched system count-rates.
• It leads to improved uniformity in reconstructed PET images.
• More uniform axial variance profiles are achieved, particularly at the scanner's axial edges.
• The method simplifies calibration by using on-the-fly acquired delayed-window data.

Conclusions:

• The novel normalization calibration method enhances quantitative PET imaging accuracy and image quality.
• It effectively mitigates common artifacts, improving diagnostic reliability.
• The simplified procedure offers practical advantages for clinical implementation in PET scans.