On the separation of quantum noise for cardiac X-ray image compression

In lossy medical image compression, the requirements for the preservation of diagnostic integrity cannot be easily formulated in terms of a perceptual model. Especially, since human visual perception is dependent on numerous factors such as the viewing conditions and psycho-visual factors. Therefore, the authors investigate the possibility to set criteria for data loss, based on the characteristics of the acquisition system. In general, medical images tend to be relatively noisy, due to the low patient dose. The nature of the noise is usually known and partially uncorrelated. In addition, the systems modulation transfer function (MTF) imposes a known spatial-frequency limitation to the information. In the assumption that noise which is not correlated with the signal contains no diagnostic information, the authors derive compression criterion based on a physical model of a digital cardiac cineangiography system