System and method for an intelligent scout scan technique to identify a high contrast object in a computed tomography scanner
A computer-implemented method includes generating an image from scan data acquired during a non-diagnostic scan utilizing a computed tomography scanner. The computer-implemented method also includes performing object segmentation based on pixel connectivity on the image to generate segmented regions. The computer-implemented method further includes characterizing pixel connectivity properties of the segmented regions. The computer-implemented method still further includes obtaining respective bounding shape coordinates and area for bounding shapes of the segmented regions.
1 . A computer-implemented method for identifying a high contrast or highly attenuating object in medical imaging data, comprising:
generating an image from scan data acquired during a non-diagnostic scan utilizing a computed tomography (CT) scanner;
performing object segmentation based on pixel connectivity on the image to generate segmented regions;
calculating pixel connectivity properties of the segmented regions, wherein the pixel connectivity properties comprise one or more of area, centroid, orientation, circularity, and eccentricity;
obtaining respective bounding shape coordinates and area for bounding shapes of the segmented regions;
designating a first bounding shape of the bounding shapes as a subject bounding shape for a subject scanned during the non-diagnostic scan;
identifying a non-overlapping region from the segmented regions having the high contrast or highly attenuating object when present;
determining whether the non-overlapping region when present is within or outside the subject bounding shape; and
generating a composite image with the subject bounding shape and a second bounding shape for the non-overlapping region when present overlaid on the image, wherein each bounding shape overlaid on the image is color-coded to indicate a status of each bounding shape, wherein the subject bounding shape is color-coded a first color to indicate that the image does not contain any high contrast or highly attenuating object and no further action is required related to any high contrast or highly attenuating object, and wherein the subject bounding shape is color coded a second color different from the first color to indicate the presence of the high contrast or highly attenuating object.
2 . The computer-implemented method of claim 1 , wherein the second bounding shape is color-coded a third color different from both the first color and the second color to indicate a presence of the high contrast or highly attenuating object outside the subject bounding shape.
3 . The computer-implemented method of claim 2 , further comprising causing display of a recommendation on a display to remove the high contrast or highly attenuating object.
4 . The computer-implemented method of claim 1 , wherein the second bounding shape is color-coded a third color different from both the first color and the second color to indicate a presence of the high contrast or highly attenuating object within the subject.
5 . The computer-implemented method of claim 4 , further comprising automatically turning on or add metal artifact reduction processing when the high contrast or highly attenuating object is present within the subject.
6 . The computer-implemented method of claim 1 , wherein the non-diagnostic scan is a scout scan.