Images processing algorithms

In the mean time, borrowing image-processing algorithms from astronomy and from high-precision photography in spy satellites, more and more visual information could be packed into a compact disk of standard audio size. 

I. Pitas, in Digital image processing algorithms and applications, Wiley, New York,... 

Quality of Image Processing Algorithms in Terms of Detectability of Lesions 43... 

From the above scene setting, it can be understood that evaluation of image processing algorithms should include both a diniccd part (evaluating the apparent global image quality) and a technical part (evaluation of contrasts and sizes of just visible lesions). 

Currently, quality assurance protocols do not provide tools to assess image processing algorithms. Algorithms could be applied repeatedly on the same set of raw clinical input images and this action could then be followed by an evaluation process. Although this would not require any additional exposure (or ethical consideration), such systematic stupes have not been reported. 

Fig. 3.16. FROC curve for one reader (Reader 1) of an observer performance experiment with simulated clusters of microcalcifications. Error bars are calculated using bootstrap method. Images were reprocessed with 2 versions of the Siemens image processing algorithms Opviewl and Opview2. Superiority of Opview 2 to Opview 1 (in terms of microcalcification visibility) is demonstrated. Figure copied from Zanca et al. (2009)...

Electronic artifacts can present as a waveform pattern (example when there is a problem with the speed of the cooling fans) or as vdiite lines, perpendicular to the pectoral side (when there is a problem with the readout of one column in the detector) (Fig. 4.13). In clinical images, these are often not visible as straight lines but as dashed lines. This is due to the interpolation of the image-processing algorithms. These artifiicts are known as vibration artifacts (Ayyala et al. 2008). 

---