Speakers
Description
Radiographic imaging is an essential diagnostic tool in medicine, but repeat exposure to radiation does not come without hazards. The acquisition of good quality images depends on the use of high energy on the X-Ray tube. Lowering that energy reduces the dose administered, however reducing X-Ray tube current and voltage can introduce Poisson noise in the images and the image exposure. With the use of Digital Radiography (DR) the images can be processed to reduce noise.
The aim of the study is to characterize the image noise in dose reduction and compare different image filtering techniques, to study possible dose reduction via DR image processing.
The noise from low dose DR images is characterized by measure of their standard deviation, and the filters are used to produce images equivalent to full dose from the lower dose scans.
The image data utilized in the study was obtained from DR chest scans of the anatomical phantom Alderson RANDO, in postanterior and lateral views. The voltage and current the images were obtained with was determined based on standard clinical doses for adult chest scans. The lower dose images were obtained at energies equivalent to pediatric chest scans, additional acquisitions were done between the two energy levels to study the progressive dose reduction.