Automatic radiography exposure control using rapid probe exposure and learned scene analysis

1. A method for capturing a diagnostic quality digital radiographic (DR) image of a subject, the method comprising:

capturing a reduced exposure DR image of a particular anatomical region of the subject in a DR detector using an x-ray source set at a reduced exposure level;

mapping one or more exposure control regions in the reduced exposure DR image based on one or more previously captured diagnostic quality DR images of the same particular anatomical region of other subjects;

calculating a difference between an exposure level used for the mapped regions in the previously captured diagnostic quality DR images and the reduced exposure level;

setting the x-ray source at a diagnostic exposure level according to the calculated difference; and

capturing the diagnostic quality DR image of the particular anatomical region of the subject in the DR detector using the x-ray source set at the diagnostic exposure level, wherein the step of setting the x-ray source includes setting an energy level of the x-ray source and a time duration of exposure by the x-ray source.

2. The method of claim 1 , wherein the step of setting the x-ray source includes determining an imaging region of the subject, a size of the subject, and anatomy view angles of the subject.

3. The method of claim 1 , wherein the step of mapping comprises receiving an output by a trained machine learning algorithm using a plurality of stored DR images.

4. The method of claim 3 , further comprising capturing the stored DR images at clinical exposure levels set at more than the reduced exposure level.

5. The method of claim 3 , further comprising training the trained machine learning algorithm using archived DR images captured at the reduced or standard diagnostic exposure levels and corresponding copies of the archived DR images having image noise induced therein to mimic exposure levels lower than standard diagnostic exposure levels.

6. The method of claim 1 , further comprising displaying the reduced exposure DR image of the subject having the one or more exposure control regions highlighted.

7. The method of claim 1 , further comprising setting the diagnostic exposure level at a higher mAs than the reduced exposure level and at a same kVp setting as the reduced exposure level.

8. The method of claim 1 , further comprising mapping boundaries of the exposure control regions to exclude non- anatomy devices in the subject.

9. A computer implemented method for capturing a diagnostic quality digital radiographic (DR) image of a subject, the method comprising:

in response to an operator command entry, capturing a reduced exposure DR image of a particular anatomy of the subject in a DR detector using an x-ray source set at a reduced exposure level;

mapping one or more exposure control regions in the reduced exposure DR image using an output from a machine learning algorithm trained on a plurality of archived diagnostic quality DR images of the same particular anatomy of other subjects;

calculating a difference between an exposure level used for the mapped regions in the plurality of archived DR images and the reduced exposure level;

automatically setting the x-ray source at a diagnostic exposure level at least about twice higher than the reduced exposure level; and

capturing the diagnostic quality DR image of the particular anatomy of the subject in the DR detector using the x-ray source set at the diagnostic exposure level.

10. The method of claim 9 , further comprising training the machine learning algorithm using the plurality of archived DR images having exposure control regions marked thereon by a human operator.

11. The method of claim 9 , further comprising training the machine learning algorithm using an iterative sequence comprising:

selecting an image from a database of the archived DR images; and

identifying at least one exposure control region in the selected image using input from a trained human observer.

12. The method of claim 9 , further comprising executing the steps of the method automatically in response to a single operator command entry.

13. The method of claim 12 , further comprising entering the single operator command entry using a pushbutton.

14. The method of claim 1 , further comprising setting the diagnostic exposure level at least about twice the reduced exposure level.

15. The method of claim 12 , further comprising executing the steps of the method automatically in less than about one- half second in response to the single operator command entry.

16. A computer implemented method for capturing a digital radiographic (DR) image of a subject, the method comprising:

in response to an operator command entry, capturing a reduced exposure DR image of the subject in a DR detector, including exposing the subject using an x-ray source set at a reduced exposure level;

mapping one or more exposure control regions in the reduced exposure DR image and associating a predetermined exposure level with the one or more exposure control regions, using an output from a machine learning algorithm;

calculating a difference between the predetermined exposure level setting and the reduced exposure level;

automatically setting the x-ray source at a diagnostic exposure level according to the calculated difference;

capturing a diagnostic DR image of the subject in the DR detector, including exposing the subject using the x-ray source set at the diagnostic exposure level; and

setting an exposure time of the x-ray source according to the calculated difference.

17. The method of claim 9 , further comprising displaying the one or more exposure control regions for confirmation by an operator before the step of calculating.

18. The method of claim 17 , further comprising performing both the steps of capturing the reduced exposure DR image and mapping one or more exposure control regions in response to a first pushbutton press by the operator, and performing the steps of calculating, automatically setting, and capturing the diagnostic quality DR image in response to a second pushbutton press by the operator.

19. The method of claim 9 , further comprising mapping one or more exposure control regions in the reduced exposure DR image to exclude man-made devices therefrom.