Deep learning “super resolution” versus iterative reconstruction: Phantom‐based image quality assessment

Key Points

• The study aims to compare the image quality improvements of deep learning super resolution against iterative reconstruction methods in phantom imaging.
• Evaluated two reconstruction techniques: PIQE 1024 and AIDR in phantom-based imaging settings.
• Conducted quantitative measurements to assess image quality and spatial resolution improvements.
• Analyzed potential for dose reduction with improved methods.
• PIQE provided substantial image quality improvements over AIDR in phantom-based quantitative measurements.
• PIQE 1024 showed improved spatial resolution compared to PIQE 512 with no apparent trade-offs in image quality.
• The study noted the potential for dose reduction when using these advanced techniques.