Patent US5982915A successfully refreshed from Google Patents.
US5982915A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Method of detecting interval changes in chest radiographs utilizing temporal subtraction combined with automated initial matching of blurred low resolution images
Assignee: Arch Development Corp
•
Filing: 1997-07-25
•
Granted: 1999-11-09
•
Expires: 2017-07-25
Abstract
Abstract A method and computerized automated initial image matching technique for enhancing detection of interval changes between temporally subsequent radiographic images via image subtraction. The method includes the steps of digitizing images, normalizing density and contrast in the digital images, correcting for lateral inclination in the digital images, detecting edges of a same feature in each image, converting the images into low resolution matrices, blurring the low resolution images, segmenting portions of the blurred low resolution matrices based on the detected edges, matching the digital images based on a cross-correlation match between the segmented portions, performing non-linear warping to further match Regions of Interest (ROI), and performing image subtraction between the matched digital images. The low resolution matrices are greater than 64x64 in size and are produced by averaging. Blurring of the low resolution matrices is performed via a Gaussian filter that removes fine structures in each image such as small vessels, bronchia, etc. The method may be performed by a computer system according to instructions stored on a computer readable medium.
Inventors
- 👤 Kunio Doi
- 👤 Takayuki Ishida
- 👤 Shigehiko Katsuragawa
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 10:53
CPC Classifications
G; G06; G06T; G06T7/00; G06T7/0002; G06T7/0012; G06T7/0014; G06T7/0016; G06T3/00; G06T3/14; G06T3/153; G06T5/00; G06T5/50; G06T7/20; G06T7/254; G06T7/30; G06T7/33; G06T7/337; G06T2207/00; G06T2207/10; G06T2207/10116; G06T2207/30; G06T2207/30004; G06T2207/30061
Claims Analysis
Total Claims: 44
Claim 1
Independent Claim
161 words
A method for detection of interval changes between first and second digital images, comprising the steps of: converting the first and second digital images to first and second low resolution images of predetermined size; blurring each low resolution image to produce first and second blurred images; correlating pixel values of the first blurred image with pixel values of the second blurred image over plural shifted portions of the first blurred image; determining a shift in position of said first blurred image which produces a maximum correlation in said correlating step and based thereon determining a global shift value representing a shift of said first digital image relative to said second digital image; producing a warped image by performing non-linear warping on one of said first and second digital images using said global shift value further to match said first and second digital images; and performing image subtraction between the warped image and the other of said first and second digital images.
Claim 2
Independent Claim
161 words
A method for detection of interval changes between first and second digital images, comprising the steps of: converting the first and second digital images to first and second low resolution images of predetermined size; blurring each low resolution image to produce first and second blurred images; correlating pixel values of the first blurred image with pixel values of the second blurred image over plural shifted portions of the first blurred image; determining a shift in position of said first blurred image which produces a maximum correlation in said correlating step and based thereon determining a global shift value representing a shift of said first digital image relative to said second digital image; producing a warped image by performing non-linear warping on one of said first and second digital images using said global shift value further to match said first and second digital images; and performing image subtraction between the warped image and the other of said first and second digital images.
Claim 3
Dependent Claim
77 words
The method according to claim 1, wherein prior to said converting step, further comprising the steps of: normalizing density and contrast between the digital images; and correcting lateral inclination present in the digital images via rotation, comprising, identifying a reference line in each digital image, determining an angle between the reference lines identified in each image, and rotating at least one of said digital images according to said angle so that said reference lines are in alignment.
Claim 4
Dependent Claim
65 words
The method according to claim 1, further comprising the steps of: detecting edges of a common feature in each of the digital images prior to said converting step; and segmenting similar portions of each blurred image based on the detected edges; wherein said step of determining a global shift value comprises the substep of determining a global shift value based on the segmented similar portions.
Claim 5
Dependent Claim
30 words
The method according to claim 1, wherein said step of converting includes the step of: converting the digital images to a low resolution matrix with a size larger than 64×64.
Claim 6
Dependent Claim
27 words
The method according to claim 1, wherein said step of converting includes the step of: reducing the digital images to a low resolution 128×128 matrix by averaging.
Claim 7
Dependent Claim
44 words
The method according to claim 1, wherein said step of blurring comprises the substeps of: inputting each low resolution image into a Gaussian filter of a predetermined matrix size; and smoothing each low resolution image by a Gaussian function implemented by said Gaussian filter.
Claim 8
Dependent Claim
54 words
The method according to claim 1, wherein said step of determining a global shift value comprises scaling coordinates that identify the maximum correlation match by a correction factor to produce said global shift value, said correction factor based on a reduction of image size occurring when said digital images are converted to low resolution.
Claim 9
Dependent Claim
123 words
The method according to claim 1, wherein said step of performing non-linear warping comprises: aligning corresponding portions of said first and second digital images, offset by said global shift value, in relation to corresponding anatomical features present in said first and second digital images; performing cross-correlation mapping to produce a plurality of shift values indicative of differences in locations between corresponding anatomical features of said first and second digital images with respect to corresponding pixel locations of said first and second digital images; fitting mathematically a predetermined surface to said shift values in order to compensate for errors introduced into said cross-correlation mapping; and shifting the anatomical features in one of said first and second digital images based on the corrected shift values.
Claim 10
Dependent Claim
30 words
The method according to claim 6, wherein said step of inputting comprises the substep of: inputting each low resolution image into a Gaussian filter of greater than 5×5 matrix size.
Claim 11
Dependent Claim
28 words
The method according to claim 6, wherein said step of inputting comprises the substep of: inputting each low resolution image into a Gaussian filter of 9×9 matrix size.
Claim 12
Dependent Claim
173 words
A computer readable medium on which is stored instructions that cause a computer system to detect interval changes between first and second digital images, by performing the steps of: converting the first and second digital images to first and second low resolution images of predetermined size; blurring each low resolution image to produce first and second blurred images, correlating pixel values of the first blurred image with pixel values of the second blurred image over plural shifted portions of the first blurred image; determining a shift in position of said first blurred image which produces a maximum correlation in said correlating step and based thereon determining a global shift value representing a shift of said first digital image relative to said second digital image; producing a warped image by performing non-linear warping on one of said first and second digital images using said global shift value further to match said first and second digital images; and performing image subtraction between the warped image and the other of said first and second digital images.
Claim 13
Dependent Claim
173 words
A computer readable medium on which is stored instructions that cause a computer system to detect interval changes between first and second digital images, by performing the steps of: converting the first and second digital images to first and second low resolution images of predetermined size; blurring each low resolution image to produce first and second blurred images, correlating pixel values of the first blurred image with pixel values of the second blurred image over plural shifted portions of the first blurred image; determining a shift in position of said first blurred image which produces a maximum correlation in said correlating step and based thereon determining a global shift value representing a shift of said first digital image relative to said second digital image; producing a warped image by performing non-linear warping on one of said first and second digital images using said global shift value further to match said first and second digital images; and performing image subtraction between the warped image and the other of said first and second digital images.
Claim 14
Dependent Claim
81 words
The computer readable medium of claim 11, wherein the stored instructions, prior to said converting step, further perform the steps of: normalizing density and contrast between the digital images; and correcting lateral inclination present in the digital images via rotation, comprising, identifying a reference line in each digital image, determining an angle between the reference lines identified in each image, and rotating at least one of said digital images according to said angle so that said reference lines are in alignment.
Claim 15
Dependent Claim
70 words
The computer readable medium of claim 11, wherein the stored instructions, further perform the steps of: detecting edges of a same feature in each of the digital images prior to said converting step; and segmenting similar portions of each blurred image based on the detected edges; wherein said step of determining a global shift value comprises the substep of determining a global shift value based on the segmented similar portions.
Claim 16
Dependent Claim
35 words
The computer readable medium of claim 11, wherein the stored instructions performing the step of converting, comprises the substep of converting the digital images to a low resolution matrices with a size larger than 64×64.
Claim 17
Dependent Claim
32 words
The computer readable medium of claim 11, wherein the stored instructions performing the step of converting, comprises the substep of reducing the digital images to a low resolution 128×128 matrix by averaging.
Claim 18
Dependent Claim
49 words
The computer readable medium of claim 11, wherein the stored instructions performing the step of blurring, comprises the substeps of: inputting each low resolution image into a Gaussian filter of a predetermined matrix size; and smoothing each low resolution image by a Gaussian function implemented by said Gaussian filter.
Claim 19
Dependent Claim
59 words
The computer readable medium of claim 11, wherein said stored instructions performing the step of determining a global shift value comprises scaling coordinates identifying the maximum correlation match by a correction factor to produce said a global shift value, said correction factor based on a reduction of image size occurring when said digital images are converted to low resolution.
Claim 20
Dependent Claim
124 words
The computer readable medium of claim 11, wherein said step of performing non-linear warping comprises: aligning corresponding portions of said first and second digital images, offset by said global shift value, in relation to corresponding anatomical features present in said first and second digital images; performing cross-correlation mapping to produce a plurality of shift values indicative of differences in locations between corresponding anatomical features of said first and second digital images with respect to corresponding pixel locations of said first and second digital images; fitting mathematically a predetermined surface to said shift values in order to compensate for errors introduced into said cross-correlation mapping; and shifting the anatomical features in one of said first and second digital images based on the corrected shift values.
Claim 21
Dependent Claim
35 words
The computer readable medium of claim 16, wherein the stored instructions performing the step of inputting, comprises the substep of inputting each low resolution image into a Gaussian filter of greater than 5×5 matrix size.
Claim 22
Dependent Claim
33 words
The computer readable medium of claim 16, wherein the stored instructions performing the step of inputting, comprises the substep of inputting each low resolution image into a Gaussian filter of 9×9 matrix size.
Claim 23
Independent Claim
170 words
A system for detection of interval changes between first and second digital images, comprising: means for converting the first and second digital images to first and second low resolution images of predetermined size; means for blurring each low resolution image to produce first and second blurred images; means for correlating pixel values of the first blurred image with pixel values of the second blurred image over plural shifted portions of the first blurred image; means for determining a shift in position of said first blurred image which produces a maximum correlation in said correlating step and based thereon determining a global shift value representing a shift of said first digital image relative to said second digital image; means for producing a warped image by performing non-linear warping on one of said first and second digital images using said global shift values further to match said first and second digital images; and means for performing image subtraction between the warped image and the other of said first and second digital images.
Claim 24
Independent Claim
170 words
A system for detection of interval changes between first and second digital images, comprising: means for converting the first and second digital images to first and second low resolution images of predetermined size; means for blurring each low resolution image to produce first and second blurred images; means for correlating pixel values of the first blurred image with pixel values of the second blurred image over plural shifted portions of the first blurred image; means for determining a shift in position of said first blurred image which produces a maximum correlation in said correlating step and based thereon determining a global shift value representing a shift of said first digital image relative to said second digital image; means for producing a warped image by performing non-linear warping on one of said first and second digital images using said global shift values further to match said first and second digital images; and means for performing image subtraction between the warped image and the other of said first and second digital images.
Claim 25
Dependent Claim
71 words
The system according to claim 21, further comprising: means for normalizing density and contrast between the digital images; means for correcting lateral inclination of the digital images, comprising, means for identifying a reference line in each digital image, means for determining an angle between the identified reference lines, and means for rotating at least one of the digital images according to said angle so that said references lines are in alignment.
Claim 26
Dependent Claim
44 words
The system according to claim 22, further comprising: means for detecting edges of a common feature in each of the digital images prior to converting by said converting means; and means for segmenting similar portions of each blurred image based on said edges detected.
Claim 27
Dependent Claim
57 words
The system according to claim 22, wherein: said means for determining a global shift value comprises means for scaling coordinates that identify the maximum correlation match by a correction factor to produce said global shift value, said correction factor based on a reduction of image size occurring when said digital images are converted to low resolution images.
Claim 28
Dependent Claim
125 words
The system according to claim 24 wherein said means for performing non-linear warping comprises: means for aligning corresponding portions of said first and second digital images in relation to corresponding anatomical features present in said first and second digital images; means for performing cross-correlation mapping to produce a plurality of shift values indicative of differences in locations between corresponding anatomical features of said first and second digital images with respect to corresponding pixel locations of said first and second digital images; means for fitting mathematically a predetermined surface to said shift values in order to compensate for errors introduced into said cross-correlation mapping; and means for shifting the anatomical features in one of said first and second digital images based on the corrected shift values.
Claim 29
Dependent Claim
29 words
The system according to claim 25, wherein said means for blurring comprises means for smoothing each low resolution image via a Gaussian filter of greater than 5×5 matrix size.
Claim 30
Dependent Claim
33 words
The system according to claim 26, wherein said means for producing low resolution images comprises means for reducing the digital images to low resolution matrices with a size greater than 64×64 by averaging.
Claim 31
Dependent Claim
24 words
The system according to claim 27, wherein: said Gaussian filter is a 9×9 Gaussian filter; and said low resolution matrices are 128×128 in size.
Claim 32
Independent Claim
101 words
A method of detecting interval changes between first and second digital images, comprising the steps of: producing a first subtraction image from said first and second digital images without initial image matching; extracting at least one objective measure of quality of said first subtraction image; evaluating said at least one objective measure of quality of said first subtraction image; and producing a second subtraction image from said first and second digital images using initial image matching only if said at least one objective measure of quality of said first subtraction image indicates that the first subtraction image is of poor quality.
Claim 33
Independent Claim
101 words
A method of detecting interval changes between first and second digital images, comprising the steps of: producing a first subtraction image from said first and second digital images without initial image matching; extracting at least one objective measure of quality of said first subtraction image; evaluating said at least one objective measure of quality of said first subtraction image; and producing a second subtraction image from said first and second digital images using initial image matching only if said at least one objective measure of quality of said first subtraction image indicates that the first subtraction image is of poor quality.
Claim 34
Dependent Claim
72 words
The method according to claim 29, further comprising the steps of: extracting at least one objective measure of quality of said second subtraction image; evaluating said at least one objective measure of quality of said second subtraction image; and producing a third subtraction image based on initially matched images from said initial image matching that includes separate and independent warping of plural selected fields present in each of the initially matched images.
Claim 35
Dependent Claim
47 words
The method according to claim 29, wherein said step of extracting comprises the substeps of: collecting average contrast and histogram data according to pixel values in at least one field of said first subtraction image; and determining if said average contrast and histogram data exceed predetermined thresholds.
Claim 36
Dependent Claim
60 words
The method according to claim 30, wherein said step of extracting at least one objective measure of quality of said second subtraction image comprises the substeps of: collecting average contrast and histogram data according to pixel values in at least one field of said second subtraction image; and determining if said average contrast and histogram data exceeds a predetermined threshold.
Claim 37
Dependent Claim
40 words
The method according to claim 30, wherein said step of producing a third subtraction image comprises: performing separate and independent warping of each of plural selected lung fields present in said initially matched images to produce a third subtraction image.
Claim 38
Dependent Claim
179 words
An image processing apparatus, comprising: a mechanism configured to convert the first and second digital images to first and second low resolution images of predetermined size; a mechanism configured to blur each low resolution image to produce first and second blurred images; a mechanism configured to correlate pixel values of the first blurred image with pixel values of the second blurred image over plural shifted portions of the first blurred image; a mechanism configured to determine a shift in position of said first blurred image which produces a maximum correlation between the second blurred image and the first blurred image and based thereon determine a global shift value representing a shift of said first digital image relative to said second digital image; a mechanism configured to produce a warped image by performing non-linear warping on one of said first and second digital images using said global shift value further to match said first and second digital images; and a mechanism configured to perform image subtraction between the warped image and the other of said first and second digital images.
Claim 39
Dependent Claim
179 words
An image processing apparatus, comprising: a mechanism configured to convert the first and second digital images to first and second low resolution images of predetermined size; a mechanism configured to blur each low resolution image to produce first and second blurred images; a mechanism configured to correlate pixel values of the first blurred image with pixel values of the second blurred image over plural shifted portions of the first blurred image; a mechanism configured to determine a shift in position of said first blurred image which produces a maximum correlation between the second blurred image and the first blurred image and based thereon determine a global shift value representing a shift of said first digital image relative to said second digital image; a mechanism configured to produce a warped image by performing non-linear warping on one of said first and second digital images using said global shift value further to match said first and second digital images; and a mechanism configured to perform image subtraction between the warped image and the other of said first and second digital images.
Claim 40
Dependent Claim
75 words
The image processing apparatus according to claim 34, further comprising: a mechanism configured to detect edges of a common feature in each of the digital images prior to said converting step; and a mechanism configured to segment similar portions of each blurred image based on the detected edges; wherein said mechanism configured to determine a global shift value further comprises a mechanism configured to determine a global shift value based on the segmented similar portions.
Claim 41
Dependent Claim
52 words
The image processing apparatus according to claim 34, wherein the mechanism configured to blur comprises: a mechanism configured to input each low resolution image into a Gaussian filter of a predetermined matrix size; and a mechanism configured to smooth each low resolution image by a Gaussian function implemented by said Gaussian filter.
Claim 42
Dependent Claim
142 words
The image processing apparatus according to claim 34, wherein the mechanism configured to perform non-linear warping comprises; a mechanism configured to align corresponding portions of said first and second digital images, offset by said global shift value, in relation to corresponding anatomical features present in said first and second digital images; a mechanism configured to perform cross-correlation mapping to produce a plurality of shift values indicative of differences in locations between corresponding anatomical features of said first and second digital images with respect to corresponding pixel locations of said first and second digital images; a mechanism configured to fit mathematically a predetermined surface to said shift values in order to compensate for errors introduced into said cross-correlation mapping; and a mechanism configured to shift the anatomical features in one of said first and second digital images based on the corrected shift values.
Claim 43
Dependent Claim
120 words
In an image processing system, an improved method of determining a shift of one digital image relative to another digital image, the improvement comprising: converting the first and second digital images to first and second low resolution images of predetermined size; blurring each low resolution image to produce first and second blurred images; correlating pixel values of the first blurred image with pixel values of the second blurred image over plural shifted portions of the first blurred image; and determining a shift in position of said first blurred image which produces a maximum correlation in said correlating step and based thereon determining a global shift value representing a shift of said first digital image relative to said second digital image.
Claim 44
Dependent Claim
120 words
In an image processing system, an improved method of determining a shift of one digital image relative to another digital image, the improvement comprising: converting the first and second digital images to first and second low resolution images of predetermined size; blurring each low resolution image to produce first and second blurred images; correlating pixel values of the first blurred image with pixel values of the second blurred image over plural shifted portions of the first blurred image; and determining a shift in position of said first blurred image which produces a maximum correlation in said correlating step and based thereon determining a global shift value representing a shift of said first digital image relative to said second digital image.
References Cited (Backward Citations)
US4710875A
Cited
US4468698A
Cited
US4644582A
Cited
US4641352A
Cited
US4899393A
Cited
US5579444A
Cited
US5179441A
Cited
US5359513A
Cited
US5604819A
Cited
US5734740A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2002-11-08 | 2003-05-10 | 2003-11-09 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2006-11-08 | 2007-05-10 | 2007-11-09 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2010-11-08 | 2011-05-10 | 2011-11-09 | $7700.00 | Window Closed / Lapsed |