Submillisievert chest CT with filtered back projection and iterative reconstruction techniques

Area of Science:

• Radiology
• Medical Imaging
• Diagnostic Imaging

Background:

• Low-dose chest CT imaging is crucial for reducing radiation exposure.
• Iterative reconstruction techniques offer potential for improving image quality at reduced doses.
• Comparing different reconstruction algorithms is essential for optimizing diagnostic performance.

Purpose of the Study:

• To compare submillisievert chest CT images reconstructed with filtered back projection (FBP), SafeCT, adaptive statistical iterative reconstruction (ASIR), and model-based iterative reconstruction (MBIR).
• To evaluate these reconstructions against standard-dose FBP images for lesion detection and visualization of anatomical structures.

Main Methods:

• Fifty patients underwent routine chest CT with an additional submillisievert series.
• Sinogram data were reconstructed using FBP, SafeCT, ASIR, and MBIR.
• Radiologists assessed lesion margins, small structure visibility, and diagnostic acceptability.
• Objective noise measurements were performed.

Main Results:

• Lesion margins were well visualized on FBP, SafeCT, and ASIR, but poorly on MBIR.
• Visibility of normal lung structures was suboptimal on MBIR.
• MBIR demonstrated the lowest image noise among the techniques.
• All submillisievert techniques were suboptimal for mediastinal structure visualization.

Conclusions:

• FBP, SafeCT, and ASIR allow optimal lesion evaluation in submillisievert chest CT.
• Submillisievert MBIR images are suboptimal for visualizing normal lung structures despite low noise.
• Further optimization is needed for mediastinal structure visualization across all submillisievert techniques.