US6310967B1 — Normal and abnormal tissue identification system …

Claim 1

Independent Claim 170 words

A method for analyzing a medical image to determine whether the image is classifiable as normal, the method comprising the steps of: applying a wavelet expansion to a digital representation of a raw image, the raw image comprising an array of sectors, each sector having an intensity level, to obtain a plurality of expansion images of varying resolution; selecting at least one expansion image having a resolution commensurate with a desired predetermined detection resolution range; dividing each expansion image into a plurality of regions, each region comprising at least one sector; and creating an output image comprising a combination of all regions for each selected expansion image, each region having a first value when the region intensity level is above a predetermined threshold level and a second value when the region intensity level is below the threshold level, for localizing a potential abnormality within the image; wherein an absence of a predetermined number of regions having a first value intensity level is indicative of the image being classifiable as normal.

Claim 2

Independent Claim 170 words

A method for analyzing a medical image to determine whether the image is classifiable as normal, the method comprising the steps of: applying a wavelet expansion to a digital representation of a raw image, the raw image comprising an array of sectors, each sector having an intensity level, to obtain a plurality of expansion images of varying resolution; selecting at least one expansion image having a resolution commensurate with a desired predetermined detection resolution range; dividing each expansion image into a plurality of regions, each region comprising at least one sector; and creating an output image comprising a combination of all regions for each selected expansion image, each region having a first value when the region intensity level is above a predetermined threshold level and a second value when the region intensity level is below the threshold level, for localizing a potential abnormality within the image; wherein an absence of a predetermined number of regions having a first value intensity level is indicative of the image being classifiable as normal.

Claim 3

Dependent Claim 119 words

The method recited in claim 1 , further comprising the steps, prior to the output image creating step, of: applying a wavelet expansion to a digital representation of a raw control image comprising an array of sectors, each sector having an intensity level, to obtain a plurality of expansion images of varying resolution; selecting at least one expansion image having a resolution commensurate with a desired predetermined detection resolution range; determining a functional form of a probability distribution function for each selected expansion image; determining an optimal statistical normal image region test for each selected expansion image; establishing the threshold level for the probability distribution function from the optimal statistical normal image region test for each selected expansion image.

Claim 4

Dependent Claim 19 words

The method recited in claim 2 , wherein the expansion-image selecting step comprises empirically selecting two adjacent expansion images.

Claim 5

Dependent Claim 57 words

The method recited in claim 3 , wherein the test-determining step comprises forming a search window having a predetermined resolution size for each selected expansion image, and wherein the output image creating step comprises combining the selected two adjacent expansion images and applying a binary mask having the predetermined resolution size to the combined expansion space images.

Claim 6

Dependent Claim 23 words

The method recited in claim 2 , wherein the optimal statistical normal image region test determining step comprises using a maximum likelihood technique.

Claim 7

Dependent Claim 30 words

The method recited in claim 2 , wherein the threshold level establishing step comprises applying an iterative procedure to determine a set of operating parameters for each selected expansion image.

Claim 8

Dependent Claim 41 words

The method recited in claim 2 , wherein the probability density function comprises one of a family of distributions having the form: P  ( ω ) = k 2  N ( k 2 + ω 2 ) N .

Claim 9

Dependent Claim 22 words

The method recited in claim 2 , wherein the test determining step comprises assuming a parametric form of a test statistic distribution.

Claim 10

Dependent Claim 29 words

The method recited in claim 1 , wherein the test determining step comprises assuming a nonparametric form of a test statistic distribution and using a kernel density estimation method.

Claim 11

Dependent Claim 31 words

The method recited in claim 1 , wherein the raw image comprises a film image, and further comprising the step, prior to the wavelet-expansion application step, of digitizing the film image.

Claim 12

Dependent Claim 21 words

The method recited in claim 1 , wherein the wavelet-expansion application step comprises performing a separable kernel two-dimensional pyramid downsampling/upsampling scheme.

Claim 13

Dependent Claim 37 words

The method recited in claim 11 , wherein each expansion image comprises a difference in information between a first image having a first resolution and a second image having a resolution one-half that of the first image.

Claim 14

Dependent Claim 27 words

The method recited in claim 1 , wherein the expansion-image selecting step comprises empirically selecting a number of expansion images commensurate with an initial raw image resolution.

Claim 15

Dependent Claim 38 words

The method recited in claim 1 , wherein the expansion image selecting step comprises selecting a plurality of expansion images, each expansion image having a resolution commensurate with a size range of an abnormality desired to be detected.

Claim 16

Dependent Claim 18 words

The method recited in claim 1 , further comprising the step of defining a tissue boundary for analysis.

Claim 17

Dependent Claim 31 words

The method recited in claim 15 , wherein the medical image comprises a digitized mammogram, and wherein the tissue boundary defining step comprises excising a breast boundary on the digitized mammogram.

Claim 18

Dependent Claim 65 words

The method recited in claim 16 , wherein the breast boundary excising step comprises the steps of: defining an off-tissue region comprising a generally random noise field and a plurality of anomalous regions; defining a tissue region comprising signal information; separating out the random noise field; determining a remaining contiguous region containing information; and setting the image region outside the remaining contiguous region to zero.

Claim 19

Dependent Claim 99 words

The method recited in claim 1 , further comprising the steps of: defining a cluster area having a size greater than a size of the region; setting a number of regions within each cluster area in the image having the first value to a first variable; comparing the first variable with a predetermined second variable; if the first variable is greater than or equal to the second variable, flagging the cluster area as potentially suspicious for the presence of an abnormality; and if the first variable is less than the second variable, classifying the cluster area to be normal.

Claim 20

Dependent Claim 33 words

The method recited in claim 1 , wherein the medical image comprises a digitized mammogram, and the expansion image selecting step comprises selecting two adjacent expansion images having resolution less than 0.5 mm.

Claim 21

Dependent Claim 21 words

The method recited in claim 19 , wherein the two selected adjacent expansion images further have resolution greater than 0.1 mm.

Claim 22

Independent Claim 147 words

An apparatus for analyzing a medical image to determine whether an abnormality is present comprising: means for applying a wavelet expansion to a digital representation of a raw image, the raw image comprising an array of sectors, each sector having an intensity level, to obtain a plurality of expansion images of varying resolution; means for selecting at least one expansion image having a resolution commensurate with a desired predetermined detection resolution range; means for dividing each expansion image into a plurality of regions, each region comprising at least one sector; and means for creating an output image comprising a combination of all regions for each selected expansion image, each region having a first value when the region intensity level is above a predetermined threshold level and a second value when the region intensity level is below the threshold level, for localizing a potential abnormality within the image.

Claim 23

Independent Claim 147 words

An apparatus for analyzing a medical image to determine whether an abnormality is present comprising: means for applying a wavelet expansion to a digital representation of a raw image, the raw image comprising an array of sectors, each sector having an intensity level, to obtain a plurality of expansion images of varying resolution; means for selecting at least one expansion image having a resolution commensurate with a desired predetermined detection resolution range; means for dividing each expansion image into a plurality of regions, each region comprising at least one sector; and means for creating an output image comprising a combination of all regions for each selected expansion image, each region having a first value when the region intensity level is above a predetermined threshold level and a second value when the region intensity level is below the threshold level, for localizing a potential abnormality within the image.

Claim 24

Dependent Claim 124 words

The apparatus recited in claim 21 , further comprising: means for applying a wavelet expansion to a digital representation of a raw control image, the raw control image comprising an array of sectors, each sector having an intensity level, to obtain a plurality of expansion images of varying resolution; means for selecting at least one expansion image having a resolution commensurate with a desired predetermined detection resolution range; means for determining a functional form of a probability distribution function for each selected expansion image; means for determining an optimal statistical normal image region test for each selected expansion image; and means for establishing the threshold level for the probability distribution function from the optimal statistical normal image region test for each selected expansion image.

Claim 25

Dependent Claim 24 words

The apparatus recited in claim 21 , wherein the raw image comprises a film image, and further comprising means for digitizing the raw image.

Claim 26

Dependent Claim 16 words

The apparatus recited in claim 21 , wherein the wavelet expansion applying means comprises software means.

Claim 27

Dependent Claim 34 words

The apparatus recited in claim 21 , wherein the output image creating means comprises software means resident within a processor and a screen in electronic communication with the processor for visualizing the output image.

Annuity Stage	Window Opens	Payment Deadline	Surcharge Period Until	Estimated Fee	Status
3.5 Years (Year 4)	2004-10-29	2005-04-30	2005-10-30	$2000.00	Window Closed / Lapsed
7.5 Years (Year 8)	2008-10-29	2009-04-30	2009-10-30	$3760.00	Window Closed / Lapsed
11.5 Years (Year 12)	2012-10-29	2013-04-30	2013-10-30	$7700.00	Window Closed / Lapsed

Normal and abnormal tissue identification system and method for medical images such as digital mammograms

Abstract

Inventors

Bibliographic Details

Claims Analysis

References Cited (Backward Citations)

USPTO Annuity & Maintenance Windows

Assigned Categories & Tags

Assign Category Tag