Computed Tomography Image Reconstruction.
Filtered back projection was used in computed tomography (CT) but produced low-dose CT images that were noisy and included artifacts. Iterative reconstruction was introduced, which reduced noise and demonstrated dose reduction; however, reconstruction times were lengthy and noise texture appeared un...
| Publicado en: | Radiologic Technology Vol. 92; no. 2; pp. 155CT - 172 |
|---|---|
| Autor principal: | |
| Formato: | CEU diagnostic images exam questions pictorial tables/charts Journal Article |
| Publicado: |
American Society of Radiologic Technologists
Nov/Dec2020
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| Sumario: | Filtered back projection was used in computed tomography (CT) but produced low-dose CT images that were noisy and included artifacts. Iterative reconstruction was introduced, which reduced noise and demonstrated dose reduction; however, reconstruction times were lengthy and noise texture appeared unnatural. Now, artificial intelligence (AI), is being applied to CT image reconstruction. These algorithms are fast and produce images comparable to those produced by iterative reconstruction. This article outlines image reconstruction techniques, including a generalized framework for deep learning. Ethics of AI in radiology also is discussed. |
|---|