Patent US4115632A successfully refreshed from Google Patents.
US5872859A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Training/optimization of computer aided detection schemes based on measures of overall image quality
Assignee: University of Pittsburgh
•
Filing: 1995-11-02
•
Granted: 1999-02-16
•
Expires: 2015-11-02
Abstract
Abstract A computerized method of detecting regions of interest in a digital image optimizes and adapts a computer aided scheme for detecting regions of interest in images. The optimization is based on global image characteristics. For each image in a database of images having known regions of interest, global image features are measured and an image characteristic index is established based on these global image features. All the images in the database are divided into a number of image groups based on the image characteristic index of each image in the database and the CAD scheme is optimized for each image group. Once the CAD scheme is optimized, to process a digital image, an image characteristics based classification criteria is established for that image, and then global image features of the digitized image are determined. The digitized image is then assigned an image characteristics rating based on the determined global image features, and the image is assigned to an image group based on the image rating. Then regions of interest depicted in the image are determined using a detection scheme adapted for the assigned image group.
Inventors
- 👤 David Gur
- 👤 Bin Zheng
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-29 12:23
CPC Classifications
G; G06; G06T; G06T7/00; G06T7/0002; G06T7/0012; G06V; G06V10/00; G06V10/20; G06V10/25; G06T2207/00; G06T2207/10; G06T2207/10116; G06T2207/30; G06T2207/30004; G06T2207/30068
Claims Analysis
Total Claims: 15
Claim 1
Independent Claim
117 words
A method of training a computer aided scheme for detecting regions of interest in images, the method comprising: (A) providing a database of images having known regions of interest; (B) for each image in the database, (b1) measuring global image features of the image in at least one of the space domain and the frequency domain, and (b2) establishing an image characteristic index for the image based on said measured global image features; (C) dividing all the images in the database into a plurality of image groups based on the image characteristic index of each image in the database; (D) training the computer aided scheme independently for each image group; and (E) establishing characteristics-based image classification criteria.
Claim 2
Independent Claim
117 words
A method of training a computer aided scheme for detecting regions of interest in images, the method comprising: (A) providing a database of images having known regions of interest; (B) for each image in the database, (b1) measuring global image features of the image in at least one of the space domain and the frequency domain, and (b2) establishing an image characteristic index for the image based on said measured global image features; (C) dividing all the images in the database into a plurality of image groups based on the image characteristic index of each image in the database; (D) training the computer aided scheme independently for each image group; and (E) establishing characteristics-based image classification criteria.
Claim 3
Dependent Claim
27 words
A method as in claim 1, wherein said images are radiographs and wherein said regions of interest are abnormal regions in living tissue depicted in said radiographs.
Claim 4
Dependent Claim
75 words
A method as in claim 2, wherein said computer aided detection scheme comprises: (a) digitizing the radiograph to produce a digitized radiograph; and, by computer, (b) identifying, according to predetermined criteria, any suspected abnormal regions; and (c) for each identified suspected abnormal region, (i) extracting multiple topographic layers of the region from the digitized radiograph; and (ii) evaluating the extracted multiple topographic layers of abnormal regions to determine whether the region is an abnormal region.
Claim 5
Dependent Claim
19 words
A method as in claim 2, wherein the radiograph is a mammogram, and wherein the abnormal regions are masses.
Claim 6
Dependent Claim
20 words
A method as in claim 2, wherein the radiograph is a mammogram, and wherein the abnormal regions are clustered microcalcifications.
Claim 7
Dependent Claim
19 words
A method as in claim 1, wherein said dividing of said images divides the images into three image groups.
Claim 8
Dependent Claim
29 words
A method as in claim 1, wherein said measuring image features for each image comprises: generating a space domain histogram of the image; and extracting features from said histogram.
Claim 9
Dependent Claim
78 words
A method as in claim 7, wherein each of said images is a radiograph depicting a breast and wherein said extracted features include at least one of: the ratio between the peak value in the histogram and the area of the breast; the difference between the peak point value and the average value of the histogram; the standard deviation of the histogram distribution; the skewness of the histogram distribution; and the number of local maxima in the histogram.
Claim 10
Dependent Claim
77 words
A method as in claim 1, wherein in the frequency domain, said features include at least one of: a ratio between the number of pixels after thresholding in the bandpass filtered images and the breast area, using a high frequency bandpass; a ratio between the number of pixels after thresholding in the bandpass filtered images and the breast area, using a low frequency bandpass; and the fractal dimension measurement in the central portion of the breast area.
Claim 11
Dependent Claim
20 words
A method as in claim 1, wherein said image characteristic index is a normalized weighted sum of said image features.
Claim 12
Dependent Claim
27 words
A method as in claim 1 wherein the computer aided scheme is a rule-based scheme and wherein the training comprises establishing different rules for each image group.
Claim 13
Dependent Claim
44 words
A method as in claim 1 wherein the computer aided scheme is a rule-based scheme and wherein the detecting of regions using a detection scheme trained specifically for the assigned image group uses a detection scheme with rules established for the assigned image group.
Claim 14
Independent Claim
212 words
A method of detecting regions of interest depicted in a digital image, the method comprising: (a) training a computer aided scheme for detecting regions of interest depicted in images, said training comprising: (i) providing a database of images having known regions of interest; (ii) for each image in the database, (A) measuring global image features in at least one of the space domain and the frequency domain, and (B) establishing an image characteristic index based on said global image features; (iii) dividing all the images in the database into a plurality of image groups based on the image characteristic index of each image in the database; (iv) optimizing the computer aided scheme independently for each image group; and (v) establishing image characteristics based classification criteria, and then (b) processing the digitized image by: (i) determining global image features of the digitized image in at least one of the space domain and the frequency domain; (ii) assigning the digitized image an image characteristics rating based on the determined global image features; (iii) assigning the image to an image group based on the image rating using the established image characteristics based classification criteria; and (iv) detecting regions of interest depicted in the image using the detection scheme optimized specifically for the assigned image group.
Claim 15
Independent Claim
212 words
A method of detecting regions of interest depicted in a digital image, the method comprising: (a) training a computer aided scheme for detecting regions of interest depicted in images, said training comprising: (i) providing a database of images having known regions of interest; (ii) for each image in the database, (A) measuring global image features in at least one of the space domain and the frequency domain, and (B) establishing an image characteristic index based on said global image features; (iii) dividing all the images in the database into a plurality of image groups based on the image characteristic index of each image in the database; (iv) optimizing the computer aided scheme independently for each image group; and (v) establishing image characteristics based classification criteria, and then (b) processing the digitized image by: (i) determining global image features of the digitized image in at least one of the space domain and the frequency domain; (ii) assigning the digitized image an image characteristics rating based on the determined global image features; (iii) assigning the image to an image group based on the image rating using the established image characteristics based classification criteria; and (iv) detecting regions of interest depicted in the image using the detection scheme optimized specifically for the assigned image group.
References Cited (Backward Citations)
US4515165A
Cited
US4907156A
Cited
US4839807A
Cited
US4851984A
Cited
US5133020A
Cited
US5212637A
Cited
US5235510A
Cited
US5369572A
Cited
US5579402A
Cited
US5289374A
Cited
US5343390A
Cited
US5319549A
Cited
US5598481A
Cited
US5627907A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2002-02-15 | 2002-08-17 | 2003-02-16 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2006-02-15 | 2006-08-17 | 2007-02-16 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2010-02-15 | 2010-08-17 | 2011-02-16 | $7700.00 | Window Closed / Lapsed |