Tomomi Takenaga, Presenter: Nothing to Disclose

Makoto Goto RT, Abstract Co-Author: Nothing to Disclose

Masahiro Hatemura, Abstract Co-Author: Nothing to Disclose

Yoshikazu Uchiyama, Abstract Co-Author: Nothing to Disclose

Shigehiko Katsuragawa PhD, Abstract Co-Author: Nothing to Disclose

Junji Shiraishi, Abstract Co-Author: Research Grant, Konica Minolta Group Research Grant, FUJIFILM Holdings Corporation Research Grant, Nihon Medi-Physics Co, Ltd

Yu Narita, Abstract Co-Author: Nothing to Disclose

Practical simulation of low-dose examinations is of immense value for optimization of CT. However, current methods are limited to specific vendor platforms, and generally rely on sinogram data that are difficult to access. We have developed a vendor-neutral computational scheme for producing simulated low-dose from standard dose CT images.

Interface: In this computer demonstration, we will present simulated low-dose clinical CT images. Examples will include brain, abdomen, and pelvic CT examinations. In a real-live presentation, any particular dose level or examination will be simulated and presented to the participant. The degree of dose reduction can be prescribed in 5% increments. The simulated CT images will be produced in real time and displayed on a side window of the standard dose CT images for comparison. Both the standard and the simulated CT images can be interpreted in scrolling mode. Methodology: Our methodology is based on adding noise to simulated sinogram data generated by Radon transform of the original CT dataset. The magnitude of noise addition is based on the difference between the tube current-exposure time product (mAs) of the original examination and the targeted reduced mAs. The approach was validated using images of the ACR CT phantom which indicated that the noise addition had the proper magnitude and texture (characterized in terms of the NPS) as those of real dose-reduced images with no negative impact on image resolution.

The simulation of low-dose CT images is a highly powerful approach to address the need for optimization in CT. However, current methods are limited to specific vendors and require access to sinogram data. Our technique eliminates these limitations with a simple methodology and a practical user interface. The approach is currently based on FBP reconstruction and needs to be extended to iterative reconstructions in the future.

We developed computer software for producing simulated low-dose CT images, which can be used for determining optimal dose setting of various CT examinations clinically without time-consuming experiments and risks in terms of increased patient dose.

http://abstract.rsna.org/uploads/2014/14009178/14009178_xzok.jpg

Takenaga, T, Goto, M, Hatemura, M, Uchiyama, Y, Katsuragawa, S, Shiraishi, J, Narita, Y, Real-time Demonstration of Simulated Low-dose Clinical CT Images.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14009178.html