The Utility of a Modified Data-Driven Motion Correction Method Incorporating Computed Tomography Positional Data for Positron Emission Tomography

Le résumé fourni par la source

In positron emission tomography/computed tomography (PET/CT), long image acquisition time often causes patient motion and misalignment between the PET and CT images. This misalignment can compromise the accuracy of attenuation correction and quantitative values, thus affecting diagnostic reliability. Recently, data-driven motion correction (DDMC) methods have been developed to detect motion using brief static PET images, and our previous study also demonstrated the effectiveness of the DDMC method. However, DDMC assumes alignment between PET reference images and CT scans used for attenuation correction. When patients move after CT scan, the PET and CT images misalign, necessitating rescanning or manual realignment. The present study introduces a modified approach, namely DDMC+CT, that integrates CT positional data into the DDMC process to address PET/CT misalignment. We evaluated the DDMC+CT approach using a phantom study and clinical images. The phantom study simulated PET/CT misalignment by moving a Hoffman phantom post-CT scan, with contrast and standardized uptake values (SUV) compared with and without DDMC+CT. The clinical images included whole-body PET/CT scans with head and neck motion, where quantitative and visual assessments were performed. The DDMC+CT method successfully corrected PET/CT misalignment in phantom and clinical images, yielding SUV and contrast values consistent with reference images, resulting in improved diagnostic accuracy. The DDMC+CT method demonstrates a potential for clinical application, where it may minimize rescans and reduce patient and operator burden by allowing automatic misalignment correction.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.