Congress:
EuroSafe Imaging 2021
Keywords:
Artificial Intelligence, Radiation physics, Radioprotection / Radiation dose, Conventional radiography, CT, Image manipulation / Reconstruction, Observer performance, Physics, Radiation safety, Quality assurance
Authors:
C. Hoeschen, G. Frija, K. Katsari, D. Koff, J. Damilakis, R. W. Loose, J. Santos, M. van Straten, I. Kralik, F. Kainberger, M. Kupinski, D. Tsetis, C. Granata, S. Schindera, D. Tack, P. Hogg
DOI:
10.26044/esi2021/ESI-15558
Conclusion
Image quality assessment is a very important tool when talking about optimising benefit/risk ratios for diagnostic imaging using ionising radiation. The iamge quality really needs to be evaluated in parallel with the dose that is applied during an imaging procedure, otherwise the dose could be reduced up to a point where relevant information is lost and thus the diagnostic performance is deteriorated. A large variety of tools is existing for the task of image quality assessment. The various different methods have been evaluated regarding their advantages and disadvantages. There is not one method that can be seen as superior to all other methods. All approaches seem to generate valuable information for different tasks. The most meaningful evaluation is probably still the subjective image quality assesmment if performed in a way that reliable results are possible. Howver, this approach can only be use for optimising an imaging procedure with respect to very few parameter settings due to the large effort necessary.
The objective image quality assessment methods vary in terms of their usability for different clinical diagnostic tasks, the amount of work related to them and their applicability with non-linear image processing approaches.