[Improvement of SNR When Adding X-ray Images with Different SNRs]

Some radiologic patient positioning techniques that can be used for X-ray examinations can be difficult to apply. One method involves using ultra-low-dose X-ray images to confirm positioning. These positioning images are typically discarded and not used for diagnosis. The purpose of this study was to improve the signal-to-noise ratio (SNR) in diagnostic imaging by including these ultra-low-dose positioning images rather than discarding them. To add two images together, we devised a method in which one image is multiplied by the coefficient calculated from its SNR before the addition. The images were dichotomized into a high SNR group and a low SNR group. The images in the high SNR group and the low SNR group were summed. When doing so, the images of the low SNR group were multiplied. There was one maximum SNR point while the SNR was being changed. The maximum SNR of the synthesized images was equal to the value of the square root of the sum of the squares of the two images. The multiplication coefficient, in the case of the maximum SNR, was near 1 when an image agreed with the Poisson distribution; when it did not, it was far from 1. The difference between the calculated values of the hypothetical measurement of the multiplication coefficient was small. In this study, we showed that improving SNR of a diagnostic image could be achieved by adding a positioning image. The multiplication coefficient in the case of the SNR maximum of a synthesized image is calculable. The measurement of a Wiener spectrum is needed for noise evaluation. There can be problems where there is motion after a positioning image is exposed.