Towards continuous-to-continuous 3D imaging in the real world

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Image Reconstruction

Background:

• Imaging systems are typically modeled as continuous-to-discrete mappings.
• Object reconstruction often assumes a discretized version, leading to discrete-to-discrete representations.
• This discretization can limit the precision of reconstructed images.

Purpose of the Study:

• To develop a single-photon emission computed tomography (SPECT) imaging method that avoids discrete representations.
• To enable image reconstruction on an arbitrarily fine set of points.
• To improve the fidelity and detail in SPECT imaging.

Main Methods:

• Developed a continuous-to-continuous imaging system model.
• Avoided discretization of the object domain.
• Avoided discretization of the imaging system representation.
• Implemented a reconstruction algorithm that operates on continuous variables.

Main Results:

• Demonstrated a method for SPECT imaging without discrete object or system representations.
• Achieved image reconstruction on an arbitrarily fine grid of points.
• The proposed method allows for enhanced detail and flexibility in SPECT reconstruction.

Conclusions:

• Continuous-to-continuous modeling offers advantages over traditional discrete approaches in SPECT.
• This method enhances the potential for high-resolution image reconstruction.
• The approach provides a more accurate representation of the underlying physical imaging process.