Patent US6137898A successfully refreshed from Google Patents.
US6137898A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Gabor filtering for improved microcalcification detection in digital mammograms
Assignee: iCad Inc
•
Filing: 1998-11-25
•
Granted: 2000-10-24
•
Expires: 2018-11-25
Abstract
Abstract A method and system for detecting and displaying clustered microcalcifications in a digital mammogram, wherein a single digital mammogram is first automatically cropped to a breast area sub-image which is then processed by means of an optimized Difference of Gaussians filter to enhance the appearance of potential microcalcifications in the sub-image. The potential microcalcifications are thresholded, clusters are detected, features are computed for the detected clusters, and the clusters are classified as either suspicious or not suspicious by means of a neural network. Thresholding is preferably by sloping local thresholding, but may also be performed by global and dual-local thresholding. The locations in the original digital mammogram of the suspicious detected clustered microcalcifications are indicated. Parameters for use in the detection and thresholding portions of the system are computer-optimized by means of a genetic algorithm. The results of the system are optimally combined with a radiologist's observation of the original mammogram by combining the observations with the results, after the radiologist has first accepted or rejected individual detections reported by the system. An alternative embodiment reduces false positive detections by means of Gabor filtering the cropped mammogram image to identify elongated structures such as milk ducts and veins. Individual microcalcifications coincident with the elongated structures are removed and the remaining detections grouped into clusters.
Inventors
- 👤 Randy P. Broussard
- 👤 Thomas F. Rathbun
- 👤 Steven K. Rogers
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-24 08:40
CPC Classifications
G; G06; G06F; G06F3/00; G06F3/14; G06F3/1415; B; B25; B25J; B25J15/00; B25J15/04; G06F18/00; G06F18/40; G06F18/41; G06T; G06T5/00; G06T5/20; G06T7/00; G06T7/0002; G06T7/0012; G06T7/10; G06T7/187; G06T7/40; G06T7/41; G06T7/42; G06V; G06V10/00; G06V10/20; G06V10/28; G06V10/40; G06V10/44; G06V10/443; A; A61; A61B; A61B6/00; A61B6/50; A61B6/502; B29; B29C; B29C2791/00; B29C2791/004; B29C2791/006; B29C2791/007; B29L; B29L2023/00; B29L2023/18; G06T2207/00; G06T2207/20; G06T2207/20081; G06T2207/20084; G06T2207/30; G06T2207/30004; G06T2207/30068; G06V2201/00; G06V2201/03
Claims Analysis
Total Claims: 24
Claim 1
Independent Claim
92 words
A method for automated detection of clustered microcalcifications from a digital mammogram comprising the steps of: obtaining a digital mammogram; detecting a first set of potential microcalcifications in said digital mammogram; Gabor filtering said digital mammogram to produce a Gabor-filtered image wherein elongated structures are identified; removing said potential microcalcifications in said first set coincident with said elongated structures to produce a second set of potential microcalcifications; detecting clusters of microcalcifications in said second set; and indicating the position in said digital mammogram of said clusters of microcalcifications in said second set.
Claim 2
Independent Claim
92 words
A method for automated detection of clustered microcalcifications from a digital mammogram comprising the steps of: obtaining a digital mammogram; detecting a first set of potential microcalcifications in said digital mammogram; Gabor filtering said digital mammogram to produce a Gabor-filtered image wherein elongated structures are identified; removing said potential microcalcifications in said first set coincident with said elongated structures to produce a second set of potential microcalcifications; detecting clusters of microcalcifications in said second set; and indicating the position in said digital mammogram of said clusters of microcalcifications in said second set.
Claim 3
Dependent Claim
84 words
Method according to claim 1 wherein said step of Gabor filtering further comprises: Gabor filtering said digital mammogram at a plurality of orientations to produce a plurality of Gabor-filtered images; producing a combined image by comparing said plurality of Gabor-filtered images and assigning, to each pixel location in said combined image, the maximum pixel value from the corresponding pixel location in each of said plurality of Gabor-filtered images; and thresholding said combined image to produce a binary image including pixels representing said elongated structures.
Claim 4
Dependent Claim
71 words
Method according to claim 2 wherein: said method further comprises, after said step of detecting a first set, and before said step of Gabor filtering, the steps of detecting clusters of microcalcifications in said first set; and identifying regions of interest including essentially only said first set clusters; and wherein said step of Gabor filtering includes Gabor filtering only areas of said digital mammogram including essentially only said regions of interest.
Claim 5
Dependent Claim
19 words
Method according to claim 3 wherein said steps of Gabor filtering and producing are repeated a plurality of times.
Claim 6
Dependent Claim
142 words
Method according to claim 4 wherein: said step of thresholding comprises for each of said regions of interest, defining a window of pixels in said digital mammogram corresponding to said region of interest; computing the mean μ(x,y) and standard deviation σ(x,y) of each said window of pixels; computing a local threshold value T(x,y) for each pixel included in each said window of pixels according to the function: T(x,y)=A+Bμ(x,y)+Cσ(x,y) where A is a predetermined offset and B and C are predetermined coefficients; comparing the gray-scale value of each said pixel included in each said window of pixels to its corresponding local threshold value; and generating said binary image by assigning a predetermined binary value to a corresponding pixel in said binary image if said gray-scale value is greater than or equal to said local threshold, and by assigning a different binary value otherwise.
Claim 7
Dependent Claim
37 words
Method according to claim 5 further comprising: before said step of thresholding, optimizing said predetermined offset A and said predetermined coefficients B and C based on statistical properties of a set of training images of digital mammograms.
Claim 8
Dependent Claim
71 words
Method according to claim 1 wherein: said method further comprises, after said step of detecting a first set, and before said step of Gabor filtering, the steps of detecting clusters of microcalcifications in said first set; and identifying regions of interest including essentially only said first set clusters; and wherein said step of Gabor filtering includes Gabor filtering only areas of said digital mammogram including essentially only said regions of interest.
Claim 9
Dependent Claim
22 words
Method according to claim 7 wherein said step of Gabor filtering further comprises Gabor filtering said Gabor-filtered image a plurality of times.
Claim 10
Dependent Claim
142 words
Method according to claim 8 wherein: said step of thresholding comprises for each of said regions of interest, defining a window of pixels in said digital mammogram corresponding to said region of interest; computing the mean μ(x,y) and standard deviation σ(x,y) of each said window of pixels; computing a local threshold value T(x,y) for each pixel included in each said window of pixels according to the function: T(x,y)=A+Bμ(x,y)+Cσ(x,y) where A is a predetermined offset and B and C are predetermined coefficients; comparing the gray-scale value of each said pixel included in each said window of pixels to its corresponding local threshold value; and generating said binary image by assigning a predetermined binary value to a corresponding pixel in said binary image if said gray-scale value is greater than or equal to said local threshold, and by assigning a different binary value otherwise.
Claim 11
Independent Claim
161 words
A method for automated clustered microcalcification detection by digital image processing in screening mammography, comprising: storing a digital representation of a mammogram; processing said digital representation to cause suspected microcalcifications to appear as bright spots in a first resulting image having a first amount of false positive suspected microcalcifications; thresholding said first resulting image to obtain a second resulting image that includes essentially only areas of suspected microcalcifications; processing said digital representation to cause elongated structures to appear as bright spots in a third resulting image; thresholding said third resulting image to obtain a fourth resulting image that includes essentially only areas of said elongated structures; and removing suspected microcalcifications from said second resulting image that correspond to said areas of said elongated structures to obtain a fifth resulting image including essentially only areas of suspected microcalcifications, said fifth resulting image having a second amount of false positive suspected microcalcifications, said second amount less than or equal to said first amount.
Claim 12
Independent Claim
161 words
A method for automated clustered microcalcification detection by digital image processing in screening mammography, comprising: storing a digital representation of a mammogram; processing said digital representation to cause suspected microcalcifications to appear as bright spots in a first resulting image having a first amount of false positive suspected microcalcifications; thresholding said first resulting image to obtain a second resulting image that includes essentially only areas of suspected microcalcifications; processing said digital representation to cause elongated structures to appear as bright spots in a third resulting image; thresholding said third resulting image to obtain a fourth resulting image that includes essentially only areas of said elongated structures; and removing suspected microcalcifications from said second resulting image that correspond to said areas of said elongated structures to obtain a fifth resulting image including essentially only areas of suspected microcalcifications, said fifth resulting image having a second amount of false positive suspected microcalcifications, said second amount less than or equal to said first amount.
Claim 13
Independent Claim
104 words
An apparatus for automated detection of clustered microcalcifications from a digital mammogram comprising: means for obtaining a digital mammogram; first detection means for detecting a first set of potential microcalcifications in said digital mammogram; Gabor filter means for filtering said digital mammogram to produce a Gabor-filtered image wherein elongated structures are identified; removal means for removing said potential microcalcifications in said first set coincident with said elongated structures to produce a second set of potential microcalcifications; second detection means for detecting clusters of microcalcifications in said second set; and indicating means for indicating the position of said clusters of microcalcifications in said digital mammogram.
Claim 14
Independent Claim
104 words
An apparatus for automated detection of clustered microcalcifications from a digital mammogram comprising: means for obtaining a digital mammogram; first detection means for detecting a first set of potential microcalcifications in said digital mammogram; Gabor filter means for filtering said digital mammogram to produce a Gabor-filtered image wherein elongated structures are identified; removal means for removing said potential microcalcifications in said first set coincident with said elongated structures to produce a second set of potential microcalcifications; second detection means for detecting clusters of microcalcifications in said second set; and indicating means for indicating the position of said clusters of microcalcifications in said digital mammogram.
Claim 15
Dependent Claim
91 words
Apparatus according to claim 11 wherein said Gabor filter means further comprises: means for Gabor filtering said digital mammogram at a plurality of orientations to produce a plurality of Gabor-filtered images; combining means for producing a combined image by comparing said plurality of Gabor-filtered images and assigning, to each pixel location in said combined image, the maximum pixel value from the corresponding pixel location in each of said plurality of Gabor-filtered images; and thresholding means for thresholding said combined image to produce a binary image including pixels representing said elongated structures.
Claim 16
Dependent Claim
62 words
Apparatus according to claim 12 wherein: said apparatus further comprises third detection means for detecting clusters of microcalcifications in said first set; and identification means for identifying regions of interest including essentially only said first set clusters; and wherein said Gabor filter means further includes means for Gabor filtering only areas of said digital mammogram including essentially only said regions of interest.
Claim 17
Dependent Claim
41 words
Apparatus according to claim 13 further comprising: means for applying said combined image, output from said combining means, to said means for Gabor filtering, for sequential processing by said means for Gabor filtering and said combining means, a plurality of times.
Claim 18
Dependent Claim
158 words
Apparatus according to claim 14 wherein: said thresholding means comprises defining means for defining, for each of said regions of interest, a window of pixels in said digital mammogram corresponding to said region of interest; first computing means for computing the mean μ(x,y) and standard deviation σ(x,y) of each said window of pixels; second computing means for computing a local threshold value T(x,y) for each pixel included in each said window of pixels according to the function: T(x,y)=A+Bμ(x,y)+Cσ(x,y) where A is a predetermined offset and B and C are predetermined coefficients; comparing means for comparing the gray-scale value of each said pixel included in each said window of pixels to its corresponding local threshold value; and generating means for generating said binary image by assigning a predetermined binary value to a corresponding pixel in said binary image if said gray-scale value is greater than or equal to said local threshold, and by assigning a different binary value otherwise.
Claim 19
Dependent Claim
41 words
Apparatus according to claim 15 further comprising: optimizing means for optimizing said predetermined offset A and said predetermined coefficients B and C, for use in said thresholding means, based on statistical properties of a set of training images of digital mammograms.
Claim 20
Dependent Claim
62 words
Apparatus according to claim 11 wherein: said apparatus further comprises third detection means for detecting clusters of microcalcifications in said first set; and identification means for identifying regions of interest including essentially only said first set clusters; and wherein said Gabor filter means further includes means for Gabor filtering only areas of said digital mammogram including essentially only said regions of interest.
Claim 21
Dependent Claim
36 words
Apparatus according to claim 17 further comprising: means for applying said Gabor-filtered image, output from said Gabor filter means, to said Gabor filter means, for sequential processing by said Gabor filter means, a plurality of times.
Claim 22
Dependent Claim
158 words
Apparatus according to claim 18 wherein: said thresholding means comprises defining means for defining, for each of said regions of interest, a window of pixels in said digital mammogram corresponding to said region of interest; first computing means for computing the mean μ(x,y) and standard deviation σ(x,y) of each said window of pixels; second computing means for computing a local threshold value T(x,y) for each pixel included in each said window of pixels according to the function: T(x,y)=A+Bμ(x,y)+Cσ(x,y) where A is a predetermined offset and B and C are predetermined coefficients; comparing means for comparing the gray-scale value of each said pixel included in each said window of pixels to its corresponding local threshold value; and generating means for generating said binary image by assigning a predetermined binary value to a corresponding pixel in said binary image if said gray-scale value is greater than or equal to said local threshold, and by assigning a different binary value otherwise.
Claim 23
Independent Claim
183 words
An apparatus for automated clustered microcalcification detection by digital image processing in screening mammography, comprising: storage means for storing a digital representation of a mammogram; first processing means for processing said digital representation to cause suspected microcalcifications to appear as bright spots in a first resulting image having a first amount of false positive suspected microcalcifications; first thresholding means for thresholding said first resulting image to obtain a second resulting image that includes essentially only areas of suspected microcalcifications; second processing means for processing said digital representation to cause elongated structures to appear as bright spots in a third resulting image; second thresholding means for thresholding said third resulting image to obtain a fourth resulting image that includes essentially only areas of said elongated structures; and removal means for removing suspected microcalcifications from said second resulting image that correspond to said areas of said elongated structures to obtain a fifth resulting image including essentially only areas of suspected microcalcifications, said fifth resulting image having a second amount of false positive suspected microcalcifications, said second amount less than or equal to said first amount.
Claim 24
Independent Claim
183 words
An apparatus for automated clustered microcalcification detection by digital image processing in screening mammography, comprising: storage means for storing a digital representation of a mammogram; first processing means for processing said digital representation to cause suspected microcalcifications to appear as bright spots in a first resulting image having a first amount of false positive suspected microcalcifications; first thresholding means for thresholding said first resulting image to obtain a second resulting image that includes essentially only areas of suspected microcalcifications; second processing means for processing said digital representation to cause elongated structures to appear as bright spots in a third resulting image; second thresholding means for thresholding said third resulting image to obtain a fourth resulting image that includes essentially only areas of said elongated structures; and removal means for removing suspected microcalcifications from said second resulting image that correspond to said areas of said elongated structures to obtain a fifth resulting image including essentially only areas of suspected microcalcifications, said fifth resulting image having a second amount of false positive suspected microcalcifications, said second amount less than or equal to said first amount.
References Cited (Backward Citations)
US4453266A
Cited
US4736439A
Cited
US4723553A
Cited
US4747156A
Cited
US4907156A
Cited
US5133020A
Cited
WO1991007135A1
Cited
US5212637A
Cited
US5463548A
Cited
US5622171A
Cited
US5668888A
Cited
US5331550A
Cited
US5291560A
Cited
US5260871A
Cited
US5343390A
Cited
US5289374A
Cited
US5740267A
Cited
US5574799A
Cited
US5625717A
Cited
US5537485A
Cited
US5267867A
Cited
US5359513A
Cited
US5661820A
Cited
US5633948A
Cited
US5365429A
Cited
US5491627A
Cited
US5729620A
Cited
US5388143A
Cited
US5452367A
Cited
US5638458A
Cited
US5825910A
Cited
US5740266A
Cited
US5598481A
Cited
US5740268A
Cited
US5666434A
Cited
US5673332A
Cited
US5768406A
Cited
US5825936A
Cited
US5769074A
Cited
US5627907A
Cited
US5579360A
Cited
US5572565A
Cited
US5761334A
Cited
US5657362A
Cited
US5586160A
Cited
US5729662A
Cited
US5857030A
Cited
US5732697A
Cited
US5757953A
Cited
US5615243A
Cited
US5799100A
Cited
US5815591A
Cited
US5768333A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2003-10-24 | 2004-04-24 | 2004-10-24 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2007-10-24 | 2008-04-24 | 2008-10-24 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2011-10-24 | 2012-04-24 | 2012-10-24 | $7700.00 | Window Closed / Lapsed |