Deblurring Algorithm for Blurred Images Caused by Patient Motion in Radiography.

Area of Science:

• Medical Imaging
• Radiography
• Image Processing

Background:

• Patient motion during radiography frequently causes blurred images, necessitating repeat scans.
• Repeat scans increase patient radiation exposure and decrease radiologic work efficiency.

Purpose of the Study:

• To propose and evaluate a novel deblurring algorithm for motion-blurred radiographic images.
• To reduce the need for repeat radiographic imaging caused by patient motion.

Main Methods:

• A video is captured during radiography using an optical device.
• Optical flow analysis between video frames estimates the point spread function (PSF).
• Image deconvolution processing is applied to restore the blurred radiographic images.

Main Results:

• The algorithm was tested on images with blur widths ranging from 1.0 mm to 5.0 mm.
• A significant reduction in blur area was observed in deblurred images.
• A significant increase in structural similarity (SSIM) was confirmed for deblurred images.

Conclusions:

• The proposed deblurring algorithm effectively restores motion-blurred radiographic images.
• This technique offers a promising solution to minimize repeat scans, radiation exposure, and improve efficiency in radiography.