Method and apparatus for expanding the use of existing computer-aided detection code

• US 7,072,498 B1
• Filed: 02/19/2002
• Issued: 07/04/2006
• Est. Priority Date: 11/21/2001
• Status: Expired due to Term

• Current Assignee: Hologic Incorporated
• Original Assignee: R2 Technology, Inc. (Hologic Incorporated)
• Inventors: Hasegawa, Akira, Zhang, Xiaolan, Chui, Haili, Roehrig, Jimmy, Pourjavid, Sussan
• Primary Examiner(s): Shah, Sanjiv
• Assistant Examiner(s): Azarian, Seyed
• Application Number:
  US10/079,327
• Time in Patent Office: 1,596 Days
• Field of Search: 382/128, 382/129, 382/130, 382/131, 382/132, 382/133, 382/162, 382/168, 382/169, 382/194, 382/196, 382/203, 382/209, 382/232, 382/234, 382/237, 382/240, 382/255, 382/274, 382/277, 382/291, 382/305, 600/408, 600/409, 378/28, 378/37, 707/102, 356/443, 345/424, 345/427
• US Class Current: 382/128
• CPC Class Codes: G06T 2207/20081 Training; Learning
• CPC Class Codes: G06T 2207/30068 Mammography; Breast
• CPC Class Codes: G06T 7/0012 Biomedical image inspection

Abstract

A method and apparatus for analyzing a medical image obtained from one of a plurality of digital modalities, the method comprising transforming or mapping the initial medical image to create a uniform contrast response and appearance regardless of the original modality of the image.

Claims

1. A system to analyze a medical image obtained using a first imaging mechanism, the system comprising an automated detector to detect abnormalities, the automated detector comprising:

automated detector code designed to detect abnormalities, the automated detector code trained using images obtained from a second imaging mechanism, the second imaging mechanism different than the first imaging mechanism used to obtain the medical image with respect to at least one of contrast response and dynamic range; and

a transform mapper to transform the medical image to have a similar contrast response and dynamic range to-images obtained with the second imaging mechanism, such that the automated detector code is applicable to images obtained with the first imaging mechanism.

2. The system of claim 1, wherein the transform mapper permits the use of an existing database of abnormalities by the automated detector code, thereby eliminating the need for separate testing and tuning of the system when the system is converted to accept images from the first imaging mechanism.

3. The system of claim 1, wherein the automated detector code is used with minor tuning of parameters, rules, and classifiers.

4. The system of claim 3, wherein the minor tuning of parameters, rules, and classifiers is done using a database of images obtained with the first imaging mechanism.

5. The system of claim 1, wherein the first imaging mechanism is a digital detector.

6. The system of claim 1, wherein the second imaging mechanism is a film-based digitized image acquisition system.

7. The system of claim 1, wherein the first imaging mechanism is a film-based digitized image acquisition system, the film used having a different response curve from the second imaging mechanism.

8. The system of claim 1, wherein the transform mapper comprises:

a lookup table to map pixels of the medical image to a standard canonical form; and

a remapper to use the lookup table to transform the medical image to the standard canonical form.

9. The system of claim 8, wherein the transform mapper further comprises:

an image origin identifier, to identify an imaging mechanism used to acquire the image; and

the lookup table including a plurality of lookup tables, each lookup table corresponding to a particular imaging mechanism; and

the remapper to select the appropriate lookup table based on the imaging mechanism identified by the image origin identifier.

10. A system to analyze medical images, the system convertible from a digitized film based system to a digitally acquired image based system, the system comprising:

a transform mapper to transform a digitally acquired image to have a contrast response and dynamic range similar to that of a standard canonical form digitized film-based image; and

computer aided diagnosis (CAD) code to detect abnormalities in the transformed digitally acquired image, the CAD code having been trained using standard canonical form digitized film based images, such that the CAD code is applicable to digitally acquired images without retraining.

11. The system of claim 10, wherein the transform mapper comprises:

a lookup table relating pixel values of the digitally acquired image to pixel values in the standard canonical form; and

a remapper to transform the digitally acquired image using the lookup table, such that the digitally acquired image corresponds to the standard canonical form.

12. The system of claim 10, wherein the transform mapper further comprises:

an image origin identifier, to identify an imaging mechanism used to acquire the image; and

the lookup table including a plurality of lookup tables, each lookup table corresponding to a particular imaging mechanism; and

the remapper to select the appropriate lookup table based on the imaging mechanism identified by the image origin identifier.

13. The system of claim 10, wherein the automated detector code is used with minor tuning of parameters, rules, and classifiers.

14. The system of claim 13, wherein the minor tuning of parameters, rules, and classifiers is done using a database of images obtained with the first imaging mechanism.

15. An apparatus for generating a transform table to convert medical images from a first format to a standard canonical format comprising:

an image acquisition system to acquire an image in the first format with a calibration design;

a step identifier to identify pixel values at various exposure levels in the image;

a mapper to map the pixel values of the image against corresponding pixel values in the standard canonical format; and

a table generator to generate a lookup table for converting pixel values from the first format to the standard canonical format, the look-up table used to transform the medical image in the first format to have a similar contrast response and dynamic range to-images in the standard canonical format, such that automated detector code is applicable to images obtained in the first format.

16. The apparatus of claim 15, further comprising a shifter to shift all pixel values by a constant to ensure that the absolute pixel values of the transformed image are in a desired range.

17. The apparatus of claim 15, wherein the calibration design is a step wedge.

18. A method to analyze medical images from a variety of sources, the method comprising:

converting a medical image acquired by one of the variety of sources to have a contrast response and dynamic range similar to that of a standard canonical form image using a lookup table; and

processing the converted medical image using computer aided diagnosis (CAD) code to detect abnormalities, the CAD code trained using standard canonical form images the CAD code being applicable to said medical image acquired by one of the variety of sources without retraining.

19. The method of claim 18, further comprising:

identifying an image source; and

retrieving an appropriate lookup table, based on the image source.

20. The method of claim 18, further comprising:

tuning the CAD code for additional precision using images from the variety of sources.

21. The method of claim 18, wherein one of the variety of sources is a digital detector.