AI-based synthetic CT attenuation correction enables reliable quantitative SPECT in unilateral condylar hyperplasia

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Artificial Intelligence in Healthcare

Background:

• Unilateral condylar hyperplasia (UCH) is a rare mandibular growth disorder requiring accurate metabolic assessment for surgical planning.
• Quantitative ^99mTc-methylene diphosphonate (MDP) SPECT/CT is standard but involves radiation exposure and potential registration errors from CT-based attenuation correction (CTAC).
• AI-generated synthetic CT (SyCT) is explored as a CT-independent alternative to reduce radiation.

Purpose of the Study:

• To evaluate the agreement between SyCT-based attenuation correction (SyCTAC) and conventional CTAC in quantitative SPECT imaging of UCH.
• To assess the clinical relevance of SyCTAC for condylar metabolic activity measurement in UCH.

Main Methods:

• Retrospective analysis of 14 UCH patients who underwent ^99mTc-MDP SPECT/CT.
• SPECT images were reconstructed using both CTAC and SyCTAC.
• Standardized uptake values were measured in the mandibular condyles and clivus; relative uptake fractions were calculated and compared using Bland-Altman analysis and linear mixed-effects models.

Main Results:

• Visual image quality was comparable between CTAC and SyCTAC reconstructions.
• Relative uptake fractions normalized to summed condylar activity showed good agreement between the two methods.
• Clivus-normalized ratios exhibited a small positive bias with SyCTAC.

Conclusions:

• AI-generated SyCT provides attenuation correction comparable to CTAC for clinically relevant relative condylar uptake assessment in UCH.
• SyCTAC supports its use as a low-radiation alternative for quantitative mandibular SPECT imaging in UCH.
• This AI-driven approach enhances diagnostic accuracy while minimizing patient radiation dose.