Low-dose CT

Computed tomography (CT) is a popular imaging method with:
Applications in biology, medicine, airport security, etc.
Despite the benefits, the widespread use of CT has raised concerns about potential cancer risk or genetic damage by X-ray radiation. Low dose CT (LDCT) can be used to minimize the risk of radiation without significantly compromising scanning or diagnostic performance.
In fact, LDCT is a widely accepted method used for two decades.
However, reducing the dose of radiation increases data noise may result in false diagnostic performance.

The main motivation is to demonstrate whether deep neural networks perform better than modern commercial iterative reconstruction methods for LDCT and to provide a basis for empowering CT reconstruction algorithms by big data and DL.
In addition to the general applicability of the image post-processing strategy, if the post-processing network performs in a way comparable or comparable to commercially available iterative algorithms, the inclusion of the machine learning elements in the field of sinograms and the reconstruction process is clear. Further Increasing the benefits of the DL approach on iterative image reconstruction.

You can follow me on Twitter, and YouTube channel