INTERACTIVE OPTIMIZATION OF SCAN DATABASES FOR STATISTICAL TESTING

• US 20150036948A1
• Filed: 01/15/2013
• Published: 02/05/2015
• Est. Priority Date: 02/07/2012
• Status: Active Grant

• Current Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Original Assignee: Koninklijke Philips N.V.
• Inventors: Wenzel, Fabian, Thiele, Frank Olaf, Young, Stewart
• Granted Patent: US 9,053,213 B2
• Time in Patent Office: Days
• US Class Current: 382/305
• CPC Class Codes: G06F 16/13 File access structures, e.g. distributed indices arrangements of input from, or output to, record carriers G06F3/06
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/583 using metadata automatically derived from the content
• CPC Class Codes: G06T 2207/30168 Image quality inspection
• CPC Class Codes: G06T 7/0014 using an image reference approach
• CPC Class Codes: G16H 30/20 for handling medical images, e.g. DICOM, HL7 or PACS
• CPC Class Codes: G16H 30/40 for processing medical images, e.g. editing

Abstract

Generating a patient image collective (34) includes receiving a plurality of candidate images (20) and associated data. At least one inclusion/exclusion rule (44) is applied to the plurality of candidate images and associated data which results in subsets of candidate images (32). The candidate images are tested which result in at least one quality measure (40). The at least one quality measure (40) and associated candidate images are reviewed. The at least one inclusion/exclusion rule (44) is refined based on the reviewed at least one quality measure (40) by at least one of: adding a rule, modifying a rule, deleting a rule, removing a candidate image; and adding a candidate image. The steps of applying the at least one inclusion/exclusion rule through refining the at least one inclusion/exclusion rule are repeated until an optimized collective of images is generated based on a collective size and the at least one quality measure (40). The generated collective is outputted to a data store (34).

Claims

1. An imaging system which generates a patient image collective, comprising:

a data store which stores a collective of images;

one or more processors programmed to;

a) receive a plurality of candidate images and associated data for inclusion in the collective of images;

b) apply at least one inclusion/exclusion rule to the plurality of candidate images and the associated data which results in a subset of candidate images;

c) test the candidate images based on at least one quality measure;

d) review the at least one quality measure and the candidate images;

e) refine the at least one inclusion/exclusion rule based on the reviewed at least one quality measure by at least one of;

adding a rule, modifying a rule, deleting a rule, removing a candidate image; and

adding a candidate image;

repeat b)-e) until an optimized collective of images is generated based on the collective size and reviewed at least one quality measure; and

output at least one generated collective to the data store or to a display.

2. The imaging system which generates the patient image collective according to claim 1, further comprising:

a workstation including;

a display connected to one or more processors which displays the at least one quality measure and the candidate images; and

at least one input device connected to the one or more processors which inputs inclusion/exclusion information from a healthcare practitioner interactively reviewing at least one of the quality measures and the candidate images.

3. The imaging system which generates the patient image collective according to claim 1, wherein the processor is further programmed to:

transform patient images to a common space; and

compare the transformed patient image to the collective of images in a voxel-wise statistical test

4. The imaging system which generates the patient image collective according to claim 1, wherein the data store includes multiple collectives.

5. The imaging system which generates the patient image collective according to claim 1, wherein the plurality of candidate images include a default collective.

6. The imaging system which generates the patient image collective according to claim 1, wherein the plurality of candidate images include collectives from different localities;

and wherein the at least one inclusion/exclusion rule includes weighting of the different locality collectives as the candidate images.

7. The imaging system according to claim 1, wherein the image collective is based on local variation.

8. A method for generating a patient image collective, comprising:

a) receiving a plurality of candidate images and associated data;

b) applying at least one inclusion/exclusion rule to the plurality of candidate images and the associated data which results in a subset of candidate images;

c) testing the candidate images based on at least one quality measure;

d) reviewing the at least one quality measure and the candidate images;

e) refining the at least one inclusion/exclusion rule based on the reviewed at least one quality measure by at least one of adding a rule, modifying a rule, deleting a rule, removing a candidate image; and

adding a candidate image;

repeating b)-e) until an optimized collective of images is generated based on a collective size and the at least one quality measure; and

outputting the generated collective to a data store or to a display

9. The method for generating the patient image collective according to claim 8, wherein there are a plurality of the inclusion/exclusion rules and further including weighting the inclusion/exclusion rules,.

10. The method for generating the patient image collective according to claim 8, wherein the plurality of candidate images and the associated data includes images and data from patient medical records.

11. The method for generating the patient image collective according to claim 8, wherein the at least one quality measure includes an outlier test in which each candidate image is left out as the remaining candidate images are checked for similarity.

12. The method for generating the patient image collective according to claim 8, wherein the at least one quality measure includes a statistical measurement of a voxel-wise subset distribution which identifies subsets of images with distributions exceeding a threshold amount.

13. The method for generating the patient image collective according to claim 8, wherein the reviewing includes generating a visual display of a statistical measurement of candidate image variation.

14. The method for generating the patient image collective according to claim 8, wherein the reviewing includes displaying an image with the image areas highlighted which exceed statistical parameters.

15. The method for generating the patient image collective according to claim 8, wherein the reviewing includes generating a visual display of representations of image subsets and at least one subset measurement,

16. The method for generating the patient image collective according to claim 8, wherein the candidate images are Positron Emission Tomography (PET) brain scans.

17. The method for generating the patient image collective according to claim 8, wherein the refining includes modifying the weight of at least one rule.

18. The method for generating the patient image collective according to claim 8, wherein the display of a statistical measurement includes a statistical polar plot.

19. A non-transient computer readable medium carrying software which controls one or more processors to perform the method of claim 8.

20. An imaging system comprising:

a data store which stores an arcane collective in a common space;

one or more processors configured to

receive candidate images,

transform the candidate images to common space,

retrieve the image collective from the data store,

refine the image collective by adding candidate images to the image collective and excluding images from the image collective by applying rules and/or quality measures based on (i) individual image characteristics, (ii) available information about patients and patient histories, (iii) clinical information from other medical examinations,