PET image reconstruction with anatomical edge guided level set prior.

Area of Science:

• Medical Imaging
• Image Reconstruction
• Computational Biology

Background:

• Positron Emission Tomography/Computed Tomography (PET/CT) systems combine anatomical and functional imaging.
• CT provides anatomical context to regularize PET images.
• Accurate localization of abnormal uptakes is crucial for disease detection.

Purpose of the Study:

• To develop a new Maximum a Posteriori (MAP) reconstruction method for PET images.
• To utilize anatomical edge information from CT to guide PET image reconstruction.
• To improve the detection and localization of abnormal uptakes in PET/CT scans.

Main Methods:

• Proposed a level set prior guided by anatomical edges from CT.
• Modeled PET image smoothness and similarity between PET and CT boundaries.
• Used Level Set Functions (LSFs) for smooth, closed functional boundaries.
• Allowed for mismatched boundaries between PET and CT data.

Main Results:

• Computer simulations demonstrated improved bias-variance performance compared to existing methods.
• The method successfully utilized incomplete anatomical edges.
• Applied to real mouse data, achieving higher contrast.
• Showed better performance in detecting simulated tumors with mismatched boundaries.

Conclusions:

• The proposed method enhances PET image reconstruction by integrating anatomical information from CT.
• It offers improved accuracy and contrast for detecting abnormalities.
• This approach is robust to boundary mismatches and incomplete edge data.