Patent US6078680A successfully refreshed from Google Patents.
US6078680A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Method, apparatus, and storage medium for detection of nodules in biological tissue using wavelet snakes to characterize features in radiographic images
Assignee: Arch Development Corp
•
Filing: 1997-07-25
•
Granted: 2000-06-20
•
Expires: 2017-07-25
Abstract
Abstract A method and apparatus for discrimination of nodules and false positives in digital chest radiographs, using a wavelet snake technique. The wavelet snake is a deformable contour designed to identify the boundary of a relatively round object. The shape of the snake is determined by a set of wavelet coefficients in a certain range of scales. Portions of the boundary of a nodule are first extracted using a multiscale edge representation. The multiscale edges are then fitted by a gradient descent procedure which deforms the shape of a wavelet snake by changing its wavelet coefficients. The degree of overlap between the fitted snake and the multiscale edges is calculated and used as a fit quality indicator for discrimination of nodules and false detections.
Inventors
- 👤 Hiroyuki Yoshida
- 👤 Shigehiko Katsuragawa
- 👤 Yali Amit
- 👤 Kunio Doi
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-29 09:38
CPC Classifications
G; G06; G06T; G06T7/00; G06T7/0002; G06T7/0012; G06V; G06V10/00; G06V10/70; G06V10/74; G06V10/75; G06V10/754
Claims Analysis
Total Claims: 41
Claim 1
Independent Claim
150 words
A method for classifying a selected feature in an original image, comprising: a) detecting at least one edge of said selected feature to make an initial determination of at least one portion of a boundary of said selected feature; b) providing a deformable contour in relation to said selected feature, the deformable contour being different from the detected at least one edge; c) overlapping the detected at least one edge with the deformable contour by deforming the deformable contour in relation to the detected at least one edge, to obtain a deformed contour approximating the boundary of said selected feature; d) determining a quality of matching in relation to a degree of overlap, resulting from step c), between the deformed deformable contour and the detected at least one edge; and e) classifying said selected feature into at least one category based on the quality of matching determined in step d).
Claim 2
Independent Claim
150 words
A method for classifying a selected feature in an original image, comprising: a) detecting at least one edge of said selected feature to make an initial determination of at least one portion of a boundary of said selected feature; b) providing a deformable contour in relation to said selected feature, the deformable contour being different from the detected at least one edge; c) overlapping the detected at least one edge with the deformable contour by deforming the deformable contour in relation to the detected at least one edge, to obtain a deformed contour approximating the boundary of said selected feature; d) determining a quality of matching in relation to a degree of overlap, resulting from step c), between the deformed deformable contour and the detected at least one edge; and e) classifying said selected feature into at least one category based on the quality of matching determined in step d).
Claim 3
Dependent Claim
44 words
A method according to claim 1, comprising: iteratively repeating steps c) and d) until said step d) indicates no further improvement in said quality of matching, whereupon said step e) of classifying is based on a maximum quality of matching determined in step d).
Claim 4
Dependent Claim
57 words
A method as recited in claim 2, wherein said step d) of determining comprises: calculating a cost function to measure said quality of matching, wherein said cost function has one of a minimum and a maximum under a condition in which an amount of overlap between said deformable contour and said at least one edge is maximized.
Claim 5
Dependent Claim
44 words
A method as recited in claim 3, wherein said step e) of classifying comprises: identifying said selected feature as one of a true nodule and a false positive, depending upon whether a value of said cost function is above a cost function threshold value.
Claim 6
Dependent Claim
37 words
A method as recited in claim 4, wherein said step a) of detecting comprises: providing a wavelet filter having a filter scale; and filtering said original image using said wavelet filter to thereby accentuate said selected feature.
Claim 7
Dependent Claim
22 words
A method as recited in claim 5, wherein said step b) of providing a deformable contour comprises providing a closed wavelet snake.
Claim 8
Dependent Claim
26 words
A method as recited in claim 6, wherein said step c) of deforming comprises: utilizing a wavelet transform to calculate a shape of said deformable contour.
Claim 9
Dependent Claim
22 words
A method as recited in claim 4, wherein said step b) of providing a deformable contour comprises providing a closed wavelet snake.
Claim 10
Dependent Claim
26 words
A method as recited in claim 8, wherein said step c) of deforming comprises: utilizing a wavelet transform to calculate a shape of said deformable contour.
Claim 11
Dependent Claim
37 words
A method as recited in claim 3, wherein said step a) of detecting comprises: providing a wavelet filter having a filter scale; and filtering said original image using said wavelet filter to thereby accentuate said selected feature.
Claim 12
Dependent Claim
22 words
A method as recited in claim 10, wherein said step b) of providing a deformable contour comprises providing a closed wavelet snake.
Claim 13
Dependent Claim
26 words
A method as recited in claim 11, wherein said step c) of deforming comprises: utilizing a wavelet transform to calculate a shape of said deformable contour.
Claim 14
Dependent Claim
22 words
A method as recited in claim 3, wherein said step b) of providing a deformable contour comprises providing a closed wavelet snake.
Claim 15
Dependent Claim
26 words
A method as recited in claim 13, wherein said step c) of deforming comprises: utilizing a wavelet transform to calculate a shape of said deformable contour.
Claim 16
Dependent Claim
44 words
A method as recited in claim 3, wherein said step e) of classifying comprises: identifying said selected feature as one of a true nodule and a false positive, depending upon whether a value of said cost function is above a cost function threshold value.
Claim 17
Dependent Claim
37 words
A method as recited in claim 2, wherein said step a) of detecting comprises: providing a wavelet filter having a filter scale; and filtering said original image using said wavelet filter to thereby accentuate said selected feature.
Claim 18
Dependent Claim
22 words
A method as recited in claim 16, wherein said step b) of providing a deformable contour comprises providing a closed wavelet snake.
Claim 19
Dependent Claim
26 words
A method as recited in claim 17, wherein said step c) of deforming comprises: utilizing a wavelet transform to calculate a shape of said deformable contour.
Claim 20
Dependent Claim
22 words
A method as recited in claim 2, wherein said step b) of providing a deformable contour comprises providing a closed wavelet snake.
Claim 21
Dependent Claim
26 words
A method as recited in claim 19, wherein said step c) of deforming comprises: utilizing a wavelet transform to calculate a shape of said deformable contour.
Claim 22
Dependent Claim
28 words
A method as recited in claim 2, wherein said step e) of classifying comprises: identifying said selected feature as one of a true nodule and a false positive.
Claim 23
Dependent Claim
57 words
A method as recited in claim 1, wherein said step d) of determining comprises: calculating a cost function to measure said quality of matching, wherein said cost function has one of a minimum and a maximum under a condition in which an amount of overlap between said deformable contour and said at least one edge is maximized.
Claim 24
Dependent Claim
44 words
A method as recited in claim 22, wherein said step e) of classifying comprises: identifying said selected feature as one of a true nodule and a false positive, depending upon whether a value of said cost function is above a cost function threshold value.
Claim 25
Dependent Claim
44 words
A method as recited in claim 22, wherein said step e) of classifying comprises: identifying said selected feature as one of a true nodule and a false positive, depending upon whether a value of said cost function is above a cost function threshold value.
Claim 26
Dependent Claim
37 words
A method as recited in claim 1, wherein said step a) of detecting comprises: providing a wavelet filter having a filter scale; and filtering said original image using said wavelet filter thereby to accentuate said selected feature.
Claim 27
Dependent Claim
22 words
A method as recited in claim 1, wherein said step b) of providing a deformable contour comprises providing a closed wavelet snake.
Claim 28
Dependent Claim
26 words
A method as recited in claim 26, wherein said step c) of deforming comprises: utilizing a wavelet transform to calculate a shape of said deformable contour.
Claim 29
Dependent Claim
34 words
A method as recited claim 26, wherein said step c) of deforming comprises: constraining a smoothness of said snake according to at least one of a minimum smoothness limit and a maximum smoothness limit.
Claim 30
Dependent Claim
28 words
A method as recited in claim 1, wherein said step e) of classifying comprises: identifying said selected feature as one of a true nodule and a false positive.
Claim 31
Dependent Claim
48 words
A method as recited in claim 1, wherein said step a) of detecting comprises: providing a first wavelet filter having a first filter scale; and filtering said original image using said first wavelet filter thereby to produce a first filtered image in which said selected feature is accentuated.
Claim 32
Dependent Claim
46 words
A method as recited in claim 30, wherein said step a) further comprises: providing a second wavelet filter having a second filter scale; filtering said original image using said second wavelet filter thereby to produce a second filtered image in which said selected feature is accentuated.
Claim 33
Dependent Claim
133 words
A method as recited in claim 31, wherein said steps c) and d) of deforming comprises: providing a first deformable contour and iteratively deforming said first deformable contour until the quality of matching between said first deformable contour and said first filtered image ceases to improve in order to determine a first maximum quality of matching; providing a second deformable contour and iteratively deforming said second deformable contour until the quality of matching between said second deformable contour and said second filtered image ceases to improve in order to determine a second maximum quality of matching; and comparing said first maximum quality of matching to said second maximum quality of matching and selecting one of said first maximum quality of matching and said second maximum quality of matching for classification in step e).
Claim 34
Dependent Claim
35 words
A method as recited claim 1, wherein said step c) of deforming comprises: constraining a smoothness of said deformable contour according to at least one of a minimum smoothness limit and a maximum smoothness limit.
Claim 35
Dependent Claim
27 words
A method as recited claim 1, wherein step a) comprises: using a multiscale edge representation of a nodule to extract portions of the boundary of the nodule.
Claim 36
Dependent Claim
127 words
The method as recited in claim 34, wherein said step of using a multiscale edge representation comprises: detecting edges of a nodule by constructing curved line segments from a filtered gradient image, including, selecting a set of local brightness maxima along a plurality of trajectories across the gradient image and selecting a set of initial edge points based on the local brightness maxima, linking adjacent edge points to form a set of curved line segments, each having a particular number of points and a particular value of average brightness, and selecting the line segments, within a predetermined range of a number of points and a predetermined range of average brightness, as a set of edge segments which approximate a border of a feature such as a nodule.
Claim 37
Dependent Claim
42 words
The method as recited in claim 35, wherein: said step b) comprises using the detected edges to estimate the entire boundary of the nodule; and said step c) comprises using a wavelet transform for modification of the shape of the deformable contour.
Claim 38
Dependent Claim
59 words
The method as recited in claim 36, wherein: said step b) of providing a deformable contour comprises providing a closed wavelet snake; and said step c) further comprises obtaining a smooth snake at each stage of deformation of said snake by restricting the range of the scales of the wavelet coefficients that define the shape of the wavelet snake.
Claim 39
Dependent Claim
49 words
The method as recited in claim 37, wherein said step e) comprises: determining a degree of overlap between the snake and the multi scale edges; and using the determined degree of overlap between the snake and the multiscale edges as a measure for classification of nodules and false positives.
Claim 40
Dependent Claim
31 words
An image processing apparatus configured to perform the steps recited in one of claims 1, 2, 22, 3, 23, 4, 25, 5, 19, 17, 27, 20, 29, 15, 33 or 37.
Claim 41
Dependent Claim
32 words
A storage medium storing a program for performing the steps recited in one of claims 1, 2, 22, 3, 23, 4, 25, 5, 19, 17, 27, 20, 29, 15, 33 or 37.
References Cited (Backward Citations)
US3882304A
Cited
US4907156A
Cited
US4839807A
Cited
US4851984A
Cited
US5579444A
Cited
US5622171A
Cited
US5463548A
Cited
US5668888A
Cited
US5289374A
Cited
US5733721A
Cited
US5319549A
Cited
US5359513A
Cited
US5487116A
Cited
US5638458A
Cited
US5598481A
Cited
US5666434A
Cited
US5627907A
Cited
US5790690A
Cited
US5768413A
Cited
US5799100A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2003-06-20 | 2003-12-20 | 2004-06-20 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2007-06-20 | 2007-12-20 | 2008-06-20 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2011-06-20 | 2011-12-20 | 2012-06-20 | $7700.00 | Window Closed / Lapsed |