System and method for performing image-based diagnosis

• US 6,115,489 A
• Filed: 09/02/1997
• Issued: 09/05/2000
• Est. Priority Date: 09/02/1997
• Status: Expired due to Fees

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Dailey, Christopher James, Afonso, Valtino Xavier, Gupta, Rajiv, Shah, Rasiklal Punjalal
• Primary Examiner(s): Au, Amelia
• Assistant Examiner(s): Bali, Vikkram
• Application Number:
  US08/921,959
• Time in Patent Office: 1,099 Days
• Field of Search: 382/152, 382/157, 382/141, 348/130, 364/474.21, 364/474.34, 702/34, 702/35
• US Class Current: 382/141
• CPC Class Codes: G06T 7/0004 Industrial image inspection
• CPC Class Codes: G16H 30/20 for handling medical images, e.g. DICOM, HL7 or PACS
• CPC Class Codes: G16H 40/40 for the management of medical equipment or devices, e.g. scheduling maintenance or upgrades
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients

Abstract

The present invention discloses a system and method for performing image-based diagnosis. In this invention, historical artifact images and corresponding actions for repairing the artifacts are acquired and stored in a database. The database of historical artifact images and corresponding actions is used to diagnose an incoming artifact image having an unknown fault.

Claims

1. A system for performing image-based diagnosis of a machine, comprising:

a database containing a plurality of historical images taken from a plurality of machines, the plurality of historical images comprising a plurality of ideal images generated from the plurality of machines using all possible machine settings and a plurality of artifact images generated from the plurality of machines, each of the artifact images having known faults associated therewith and a corresponding corrective action for repairing the faults; and

a diagnostic unit for diagnosing a new artifact image from a machine having an unknown fault, the diagnostic unit comprising a diagnostic image processor comprising means for finding an ideal image from the plurality of historical images that most closely matches the new artifact image, means for assigning an artifact category to the new artifact image based on the matched ideal image, and means for extracting an artifact feature from the new artifact image according to the assigned category; and

a diagnostic fault isolator comprising means for generating a plurality of metrics for the extracted artifact feature and means for applying the plurality of metrics to identify an artifact image from the plurality of historical images that most closely matches the new artifact image and a corrective action for repairing the unknown fault.

2. The system according to claim 1, further comprising a training unit coupled to the database and the diagnostic unit, the training unit comprising a training image processor comprising means for obtaining the plurality of artifact images and the plurality of ideal images, means for matching each of the plurality of artifact images to a corresponding ideal image, means for assigning an artifact category to each match, and means for extracting an artifact feature from each match according to the assigned category.

3. The system according to claim 2, wherein the training image processor further comprises means for registering each of the plurality of artifact images to its matched corresponding ideal image and means for removing each of the corresponding ideal images from its matched artifact image after registration.

4. The system according to claim 2, wherein the training unit further comprises a training fault isolator coupled to the training image processor for isolating the extracted artifact features into historical cases.

5. The system according to claim 4, wherein the database further comprises a plurality of error logs generated from the plurality of machines, each of the plurality of error logs containing data representative of events occurring during operation of the machines.

6. The system according to claim 5, wherein the training fault isolator combines the extracted artifact features and error logs into historical cases.

7. The system according to claim 1, wherein the diagnostic image processor further comprises means for registering the new artifact image to the ideal image that most closely matches the new artifact image and means for removing the corresponding ideal image from the new artifact image after registration.

8. The system according to claim 1, wherein the diagnostic fault isolator comprises means for receiving an error log generated from the machine having the unknown fault, the error log containing data representative of events occurring during operation of the machine.

9. The system according to claim 8, wherein the diagnostic fault isolator uses the error log to generate the plurality of metrics.

10. The system according to claim 1, wherein the diagnostic unit further comprises means for adding newly identified artifact images and corresponding corrective actions to the plurality of artifact images in the database.

11. A method for performing image-based diagnosis of a machine, comprising the steps of:

obtaining a plurality of historical images taken from a plurality of machines, the plurality of historical images comprising a plurality of ideal images generated from the plurality of machines using all possible machine settings and a plurality of artifact images generated from the plurality of machines, each of the artifact images having known faults associated therewith and a corresponding corrective action for repairing the faults;

receiving a new artifact image from a machine having an unknown fault;

finding an ideal image from the plurality of historical images that most closely matches the new artifact image;

assigning an artifact category to the new artifact image based on the ideal image that most closely matches the new artifact image;

generating a plurality of metrics for the artifact category assigned to the new artifact image; and

using the plurality of metrics to identify an artifact image from the plurality of historical images that most closely matches the new artifact image and a corrective action for repairing the unknown fault.

12. The method according to claim 11, wherein the step of obtaining the plurality of historical images comprises the steps of:

matching each of the plurality of artifact images to a corresponding ideal image taken from the plurality of ideal images;

assigning an artifact category to each match; and

extracting an artifact feature from each match.

13. The method according to claim 12, further comprising the steps of:

registering the artifact image with its known fault to the corresponding ideal image; and

removing the corresponding ideal image from the registered image.

14. The method according to claim 11, further comprising the step of determining an artifact feature for the new artifact image.

15. The method according to claim 11, further comprising the steps of:

registering the new artifact image to a corresponding ideal image; and

removing the corresponding ideal image from the registered image.

16. The method according to claim 11, further comprising the step of adding newly identified artifact images and corresponding corrective actions to the plurality of artifact images in the plurality of historical images.

17. The method according to claim 11, further comprising the step of inputting an error log from the machine having the unknown fault, the error log containing data representative of events occurring during operation of the machines.

18. The method according to claim 17, wherein the inputted error log is used to generate the plurality of metrics for new artifact image.