In a method, tomography apparatus and software product for evaluating time varying diagnostic images, a series of images are obtained in a time sequence from an examination subject in a tomography system, the sequence including exposures in a first image dataset obtained before occurrence of a diagnostically relevant event associated with the subject, and a second image dataset obtained at or after the event. A first correction of the image data is undertaken by displacing the image contents in each image to bring the position of the subject in each image into coincidence. A second correction of the image data is undertaken by obtaining an average value image from the image data corrected by the first correction. All of the images in the first and second datasets, corrected by the first correction, are normalized to the average value image on a picture element-by-picture element basis. A third correction is undertaken wherein the value of a picture element of a structurally uniform image region in an image in the first image dataset, corrected according to the second correction, or an average of picture elements in this image region in the first image dataset, is subtracted from the respective values of all picture elements in this region in the first and second image datasets.

 
Web www.patentalert.com

> Method and apparatus for determining a region in an image based on a user input

~ 00362