Labeling system for cross-sectional medical imaging examinations
This patent includes a method for displaying a reference image to the radiologist similar to the current image the radiologist is actively reviewing. Additionally, this patent provides a method to enhance both an educational experience and an image analysis process for an imaging examination by incorporating classification of anatomic features and methods to teach a user the names of imaging findings.
1. A method comprising:
determining at least one unlabeled structure within a cross-sectional medical imaging examination
wherein said at least one unlabeled structure comprises imaging feature(s) within said cross-sectional medical imaging examination,
wherein said at least one unlabeled structures comprises an anatomic finding,
wherein said at least one unlabeled structure does not have an associated text label(s), and
wherein said cross-sectional medical imaging examination comprises at least one of a computed tomography (CT) scan, a magnetic resonance imaging (MRI) examination, a positron emission tomography (PET) scan, a single photon emission computed tomography (SPECT) scan and an ultrasound examination;
performing an analysis of said at least one unlabeled structure
wherein said analysis comprises an artificial intelligence (AI) algorithm,
wherein said analysis determines a text label for each unlabeled structure of said at least one unlabeled structure to cause said at least one unlabeled structure to become at least one labeled structure; and
presenting a labeled cross-sectional imaging examination to a user wherein said labeled cross-sectional imaging examination contains said at least one labeled structure.
2. The method of claim 1 further comprising wherein said determining said at least one unlabeled structure is based on selection by said user.
3. The method of claim 1 further comprising wherein said determining said at least one unlabeled structure is based on a second AI algorithm.
4. The method of claim 1 further comprising wherein an optimized reference image is presented adjacent to said labeled cross-sectional imaging examination.
5. The method of claim 1 further comprising wherein an optimized reference image is presented superimposed on said labeled cross-sectional imaging examination.
6. The method of claim 1 further comprising wherein said at least one unlabeled structures comprises a pathologic finding.
7. The method of claim 1 further comprising wherein said at least one unlabeled structures comprises a surgical device.
8. The method of claim 1 further comprising wherein said at least one unlabeled structures comprises a medical device.
9. The method of claim 1 further comprising wherein said at least one unlabeled structures comprises an artifact.
10. The method of claim 1 further comprising wherein said at least one unlabeled structures comprises a foreign body.
11. The method of claim 1 further comprising wherein said at least one unlabeled structures comprises a feature identified as abnormal on a prior imaging examination.
12. The method of claim 1 further comprising wherein said at least one unlabeled structures comprises an imaging feature known to be poorly understood by said user.
13. The method of claim 1 further comprising presenting a location indicator at the at least one labeled structure to communicate to said user the precise spot of the label on the image.
14. The method of claim 13 further comprising wherein said location indicator comprises a digital object placed at the site of the at least one labeled structure.
15. The method of claim 13 further comprising:
wherein said location indicator comprises a cursor hovering over structure of interest, and
wherein said label is displayed on a monitor.
16. The method of claim 13 further comprising wherein said location indicator is a line to connect said at least one labeled structure of interest to a label.
17. The method of claim 1 further comprising wherein said determining at least one unlabeled structure within a cross-sectional medical imaging examination is based on eye tracking of said user.
18. A method comprising:
presenting at least one unlabeled structure within a cross-sectional medical imaging examination
wherein said at least one unlabeled structure is selected by a first artificial intelligence (AI) algorithm,
wherein said first AI algorithm classifies said structure as abnormal,
wherein said at least one unlabeled structure comprises imaging feature(s) within said cross-sectional medical imaging examination,
wherein said at least one unlabeled structure does not have an associated text label(s), and
wherein said cross-sectional medical imaging examination comprises at least one of a computed tomography (CT) scan, a magnetic resonance imaging (MRI) examination, a positron emission tomography (PET) scan, a single photon emission computed tomography (SPECT) scan and an ultrasound examination;
performing an analysis of said at least one unlabeled structure
wherein said analysis comprises a second AI algorithm,
wherein said analysis assigns a text label for each unlabeled structure of said at least one unlabeled structure to cause said at least one unlabeled structure to become at least one labeled structure;
presenting a labeled cross-sectional imaging examination to a user wherein said labeled cross-sectional imaging examination contains said at least one labeled structure; and
presenting a location indicator at the at least one labeled structure to communicate to said user a precise spot of a label on an image wherein said location indicator comprises a line connecting said at least one labeled structure to said label.
19. A method comprising:
loading a cross-sectional medical imaging examination into an image processing system
wherein at least one unlabeled structure comprises imaging feature(s) within said cross-sectional medical imaging examination,
wherein said at least one unlabeled structure within said cross-sectional medical imaging examination is determined based on eye tracking of a user,
wherein said at least one unlabeled structure does not have an associated text label(s), and
wherein said cross-sectional medical imaging examination comprises at least one of a computed tomography (CT) scan, a magnetic resonance imaging (MRI) examination, a positron emission tomography (PET) scan, a single photon emission computed tomography (SPECT) scan and an ultrasound examination;
performing an analysis of said at least one unlabeled structure by said image processing system
wherein said analysis comprises artificial intelligence,
wherein said analysis assigns a text label for each unlabeled structure of said at least one unlabeled structure to cause said at least one unlabeled structure to become at least one labeled structure;
presenting a labeled cross-sectional imaging examination wherein said labeled cross-sectional imaging examination contains said at least one labeled structure.