TU-E-217BCD-10: Dose Reduction in Digital Breast Tomosynthesis with the Dose Reduction Prior Image Constrained...

Area of Science:

• Medical Imaging
• Radiology
• Image Processing

Background:

• Digital Breast Tomosynthesis (DBT) is crucial for breast cancer detection.
• Image noise and radiation dose are significant concerns in DBT.
• Prior Image Constrained Compressed Sensing (PICCS) is an advanced reconstruction technique.

Purpose of the Study:

• To evaluate the effectiveness of the PICCS algorithm in reducing image noise and radiation dose for DBT.
• To assess the impact of PICCS on the diagnostic quality of reconstructed breast images.

Main Methods:

• Retrospective analysis of clinical DBT datasets using Hologic Selenia Dimensions systems.
• Prospective study with an American College of Radiology breast phantom at varying radiation doses.
• Application of the PICCS algorithm for image reconstruction and noise reduction.
• Visual inspection and quantitative measurement of noise levels in reconstructed images.

Main Results:

• PICCS reconstructions demonstrated over 35% noise reduction in clinical data.
• No loss of critical diagnostic features (calcifications, low contrast lesions) was observed; visibility of low contrast lesions improved.
• A 28% radiation dose reduction was achieved with phantom data.
• Low-dose PICCS reconstructions of phantom data showed enhanced low contrast lesion detectability and reduced noise.

Conclusions:

• The PICCS algorithm shows potential for significant dose savings in DBT.
• Diagnostic image quality at a 28% reduced dose was equivalent to or superior to full-dose acquisitions.
• Noise suppression via PICCS enhances the visibility of low contrast lesions in DBT without compromising essential diagnostic information.