Effect of pixel size on detectability of low-contrast signals in digital radiography

The effect of pixel size and other physical parameters on the detectability of simple signals in digital radiography was investigated using a signal-to-noise ratio (SNR) that is based on statistical decision theory and takes into account the characteristics of the human observer. The calculation of the SNR includes two-dimensional displayed digital signal spectra and noise Wiener spectra as well as the visual-response function and internal noise of the human observer. Threshold contrasts derived from calculated SNR's agreed well with results of an 18-alternative forced-choice observer study. Since a digital imaging system is not shift invariant, the threshold contrasts were determined for a range of alignments of the object relative to the sampling coordinates. The results indicate that a 0.2-mm pixel size may be adequate for visual detection of necessary detail in most digital radiographic examinations.