Next generation tomosynthesis image acquisition optimization for dedicated PET-DBT attenuation corrections.

Area of Science:

• Medical Imaging
• Radiology
• Biomedical Engineering

Background:

• Digital breast tomosynthesis (DBT) is a key imaging modality for breast cancer detection.
• Integrating positron emission tomography (PET) with DBT (PET-DBT) offers potential for improved diagnostic accuracy.
• Optimizing acquisition geometries is crucial for enhancing PET-DBT performance.

Purpose of the Study:

• To investigate custom acquisition geometries for a next-generation tomosynthesis (NGT) prototype.
• To improve dense/adipose tissue classification within the breast.
• To enhance breast outline segmentation for better image analysis.

Main Methods:

• Simulated various NGT acquisition geometries using the OpenVCT virtual clinical trial framework.
• Employed anthropomorphic breast phantoms for realistic simulations.
• Compared custom NGT geometries (40-200 mm PA motion) against conventional DBT.
• Utilized receiver operating characteristic (ROC) curves and area under the curve (AUC) to quantify classification performance.
• Calculated reconstructed breast volume and sensitivity index (d').

Main Results:

• Increasing posteroanterior (PA) x-ray source motion in NGT reduced breast volume overestimation compared to conventional DBT.
• Both AUC and sensitivity index (d') increased with greater PA source motion.
• Custom NGT geometries demonstrated superior adipose-glandular classification and segmentation.

Conclusions:

• Next-generation tomosynthesis (NGT) acquisition geometries can significantly improve breast tissue classification and segmentation.
• Optimized NGT parameters enhance image quality for PET-DBT applications.
• These findings suggest NGT offers improved attenuation corrections for PET-DBT over conventional DBT.