Structural similarity analysis of midfacial fractures-a feasibility study

Area of Science:

• Medical Imaging
• Radiology
• Biomedical Engineering

Background:

• Computed tomography (CT) is crucial for diagnosing midfacial fractures.
• Assessing fracture characteristics and image quality is vital for treatment planning.
• Reducing radiation dose in CT scans is a significant clinical goal.

Purpose of the Study:

• To evaluate the feasibility of the structural similarity index metric (SSIM) for analyzing midfacial fractures.
• To assess the impact of radiation dose reduction, iterative reconstruction (IR), and deep learning reconstruction on CT image quality and fracture characteristics.

Main Methods:

• Zygomaticomaxillary fractures were created in human cadaver specimens.
• Scans were performed using standard and low-dose CT protocols.
• Datasets were reconstructed with varying IR strengths and processed with a deep learning algorithm (PixelShine™).
• SSIM was used to measure structural image quality and compare fracture characteristics to contralateral anatomy.

Main Results:

• Radiation dose reduction significantly impacted structural image quality.
• Iterative reconstruction strength had a smaller effect on image quality but influenced fracture characteristics.
• The deep learning algorithm had minimal impact on both structural image quality and fracture characteristics.
• SSIM proved feasible for analyzing structural image quality and fracture characteristics.

Conclusions:

• The structural similarity index metric is a viable tool for assessing CT image quality and midfacial fracture characteristics.
• While radiation dose reduction and IR affect image analysis, deep learning reconstruction shows promise in preserving structural integrity.
• Further research can optimize CT protocols for fracture assessment with reduced radiation exposure.