US6026174A — System and method for automatically detecting mal…

Claim 1

Independent Claim 141 words

A method for detecting epithelial cells in a cell sample, comprising the steps of: a. obtaining a cell sample; b. fixing the cells of the cell sample; c. staining the cells to identify cell nuclei in the cell sample; d. illuminating the sample and obtaining an image of the sample with a microscope and a digital camera; e. compensating the image for variations in background illumination; f. analyzing the image to detect objects of interest; g. determining a focus setting for each object of interest and obtaining an image of each object of interest at its determined focus setting; h. calculating an edge that bounds each object of interest; i. calculating a set of feature values for each object of interest; and j. providing the set of feature values to a classifier that identifies epithelial cells in the objects of interest.

Claim 2

Independent Claim 141 words

A method for detecting epithelial cells in a cell sample, comprising the steps of: a. obtaining a cell sample; b. fixing the cells of the cell sample; c. staining the cells to identify cell nuclei in the cell sample; d. illuminating the sample and obtaining an image of the sample with a microscope and a digital camera; e. compensating the image for variations in background illumination; f. analyzing the image to detect objects of interest; g. determining a focus setting for each object of interest and obtaining an image of each object of interest at its determined focus setting; h. calculating an edge that bounds each object of interest; i. calculating a set of feature values for each object of interest; and j. providing the set of feature values to a classifier that identifies epithelial cells in the objects of interest.

Claim 3

Dependent Claim 108 words

The method of claim 1, wherein the features used to identify epithelial cells in the cell sample comprise features selected from the group consisting of area, mean radius, OD variance, OD skewness, range average, OD maximum, density of light spots, low DNA area, high DNA area, low DNA amount, high DNA amount, high average distance, mid/high average distance, correlation, homogeneity, entropy, fractal dimension, DNA index, run 0 percent, run 45 percent, run 90 percent, run 135 percent, grey level 0, grey level 45, grey level 90, grey level 135, run length 0, run length 45, run length 90, run length 135, harmonic 4, harmonic 5, and harmonic 6.

Claim 4

Dependent Claim 30 words

The method of claim 1, wherein the classifier identifies epithelial cells using a discriminant function, wherein the discriminant function uses features selected from the group consisting of harmon05 and freqmac2.

Claim 5

Dependent Claim 14 words

The method of claim 1, wherein the cell sample is a human lung specimen.

Claim 6

Dependent Claim 20 words

The method of claim 1, wherein staining the sample to identify cell nuclei comprises staining with a stoichiometric DNA stain.

Claim 7

Dependent Claim 28 words

The method of claim 5, wherein the stoichiometric DNA stain is selected from the group consisting of a Feulgen stain, a Romanowski stain, May-Grunwald-Giemsa stain, and Methyl Green.

Claim 8

Dependent Claim 12 words

The method of claim 5, wherein the stoichiometric DNA stain is thionin.

Claim 9

Independent Claim 143 words

A method for detecting a diagnostic cell in a cell sample, comprising the steps of: a. obtaining a cell sample; b. fixing the cells of the cell sample; c. staining the cells to identify cell nuclei in the cell sample; d. illuminating the sample and obtaining an image of the sample with a microscope and a digital camera; e. compensating the image for variations in background illumination; f. analyzing the image to detect objects of interest; g. determining a focus setting for each object of interest and obtaining an image of each object of interest at its determined focus setting; h. calculating an edge that bounds each object of interest; i. calculating a set of feature values for each object of interest; and j. providing the set of feature values to a classifier that identifies a diagnostic cell in the objects of interest.

Claim 10

Independent Claim 143 words

A method for detecting a diagnostic cell in a cell sample, comprising the steps of: a. obtaining a cell sample; b. fixing the cells of the cell sample; c. staining the cells to identify cell nuclei in the cell sample; d. illuminating the sample and obtaining an image of the sample with a microscope and a digital camera; e. compensating the image for variations in background illumination; f. analyzing the image to detect objects of interest; g. determining a focus setting for each object of interest and obtaining an image of each object of interest at its determined focus setting; h. calculating an edge that bounds each object of interest; i. calculating a set of feature values for each object of interest; and j. providing the set of feature values to a classifier that identifies a diagnostic cell in the objects of interest.

Claim 11

Dependent Claim 105 words

The method of claim 8, wherein the features used to identify a diagnostic cell comprise features selected from the group consisting of area, mean radius, OD variance, OD skewness, range average, OD maximum, density of light spots, low DNA area, high DNA area, low DNA amount, high DNA amount, high average distance, mid/high average distance, correlation, homogeneity, entropy, fractal dimension, DNA index, run 0 percent, run 45 percent, run 90 percent, run 135 percent, grey level 0, grey level 45, grey level 90, grey level 135, run length 0, run length 45, run length 90, run length 135, harmonic 4, harmonic 5, and harmonic 6.

Claim 12

Dependent Claim 52 words

The method of claim 8, wherein the features used to identify a diagnostic cell comprise features selected from the group consisting of area, density of light spots, low DNA area, high DNA area, low DNA amount, high DNA amount, correlation, homogeneity, entropy, fractal dimension, DNA index, OD maximum, and medium DNA amount.

Claim 13

Dependent Claim 42 words

The method of claim 8, wherein the classifier identifies a diagnostic cell in a cell sample using a discriminant function, wherein the discriminant function uses features selected from the group consisting of harmon03 fft, cl shade, den drk spot, and fractal2 area.

Claim 14

Dependent Claim 13 words

The method of claim 8, wherein the diagnostic cell is diagnostic of cancer.

Claim 15

Dependent Claim 14 words

The method of claim 10, wherein the diagnostic cell is a preinvasive cancerous cell.

Claim 16

Dependent Claim 14 words

The method of claim 10, wherein the diagnostic cell is an invasive cancerous cell.

Claim 17

Independent Claim 177 words

A method for screening a patient for cancer, comprising the steps of: a. obtaining a cell sample; b. fixing the cells of the cell sample; c. staining the cells to identify cell nuclei in the cell sample; d. illuminating the sample and obtaining an image of the sample with a microscope and a digital camera; e. compensating the image for variations in background illumination; f. analyzing the image to detect objects of interest; g. determining a focus setting for each object of interest and obtaining an image of each object of interest at its determined focus setting; h. calculating an edge that bounds each object of interest; i. calculating a set of feature values for each object of interest; j. providing the set of feature values to a first classifier that identifies epithelial cells in the objects of interest; and k. providing the set of feature values calculated for the objects of interest that were identified as epithelial cells to a second classifier that identifies whether the epithelial cells include diagnostic cells in the objects of interest.

Claim 18

Independent Claim 177 words

A method for screening a patient for cancer, comprising the steps of: a. obtaining a cell sample; b. fixing the cells of the cell sample; c. staining the cells to identify cell nuclei in the cell sample; d. illuminating the sample and obtaining an image of the sample with a microscope and a digital camera; e. compensating the image for variations in background illumination; f. analyzing the image to detect objects of interest; g. determining a focus setting for each object of interest and obtaining an image of each object of interest at its determined focus setting; h. calculating an edge that bounds each object of interest; i. calculating a set of feature values for each object of interest; j. providing the set of feature values to a first classifier that identifies epithelial cells in the objects of interest; and k. providing the set of feature values calculated for the objects of interest that were identified as epithelial cells to a second classifier that identifies whether the epithelial cells include diagnostic cells in the objects of interest.

Claim 19

Independent Claim 129 words

A method for determining whether a patient will develop invasive cancer, comprising the steps of: obtaining a cell sample from the patient; determining whether the cells in the sample include a diagnostic cell by: (1) staining the nuclei of the cells in the sample; (2) obtaining an image of the cells with a digital microscope and recording the image in a computer system; (3) analyzing the stored image of the cells to identify epithelial cells; (4) computing a set of feature values for the epithelial cells identified in the sample and from the feature values determining whether the epithelial cells include a diagnostic cell; and determining a total number of diagnostic cells in the cell sample and from the total number predicting whether the patient will develop invasive cancer.

Claim 20

Independent Claim 129 words

A method for determining whether a patient will develop invasive cancer, comprising the steps of: obtaining a cell sample from the patient; determining whether the cells in the sample include a diagnostic cell by: (1) staining the nuclei of the cells in the sample; (2) obtaining an image of the cells with a digital microscope and recording the image in a computer system; (3) analyzing the stored image of the cells to identify epithelial cells; (4) computing a set of feature values for the epithelial cells identified in the sample and from the feature values determining whether the epithelial cells include a diagnostic cell; and determining a total number of diagnostic cells in the cell sample and from the total number predicting whether the patient will develop invasive cancer.

Claim 21

Dependent Claim 13 words

The method of claim 16, wherein the invasive cancer is an epithelial cancer.

Claim 22

Dependent Claim 30 words

The method of claim 17, wherein the epithelial cancer is selected from the group consisting of lung cancer, breast cancer, prostate cancer, skin cancer, and cancer of the gastrointestinal tract.

Claim 23

Dependent Claim 158 words

An automated cytological specimen classifier for identifying diagnostic cells, comprising: a microscope for obtaining a view of a cytological specimen located on a slide; a camera for creating an image of the view; an image digitizer for producing a digital representation of the image; and a computer system for controlling and interfacing the microscope, camera, and image digitizer, wherein the computer system analyzes the digital representation of the image to locate one or more objects of interest and calculates a set of feature values for each object of interest, the computer system further including: a first classifier for identifying normal and abnormal epithelial cells in the digital representation of the image based on a first set of feature values computed for the object of interest; and a second classifier for identifying normal epithelial cells as diagnostic cells based on a second set of feature values computed for the objects of interest that were identified as normal epithelial cells.

Claim 24

Dependent Claim 158 words

An automated cytological specimen classifier for identifying diagnostic cells, comprising: a microscope for obtaining a view of a cytological specimen located on a slide; a camera for creating an image of the view; an image digitizer for producing a digital representation of the image; and a computer system for controlling and interfacing the microscope, camera, and image digitizer, wherein the computer system analyzes the digital representation of the image to locate one or more objects of interest and calculates a set of feature values for each object of interest, the computer system further including: a first classifier for identifying normal and abnormal epithelial cells in the digital representation of the image based on a first set of feature values computed for the object of interest; and a second classifier for identifying normal epithelial cells as diagnostic cells based on a second set of feature values computed for the objects of interest that were identified as normal epithelial cells.

Claim 25

Dependent Claim 13 words

The automated classifier of claim 19 wherein the microscope is a digital microscope.

Claim 26

Dependent Claim 13 words

The automated classifier of claim 19 wherein the camera is a CCD camera.

Claim 27

Dependent Claim 105 words

The automated classifier of claim 19 wherein the first and second set of feature values are selected from the group consisting of area, mean radius, OD variance, OD skewness, range average, OD maximum, density of light spots, low DNA area, high DNA area, low DNA amount, high DNA amount, high average distance, mid/high average distance, correlation, homogeneity, entropy, fractal dimension, DNA index, run 0 percent, run 45 percent, run 90 percent, run 135 percent, grey level 0, grey level 45, grey level 90, grey level 135, run length 0, run length 45, run length 90, run length 135, harmonic 4, harmonic 5, and harmonic 6.

Claim 28

Dependent Claim 49 words

The automated classifier of claim 19 wherein the second detectable features comprise features selected from the group consisting of area, density of light spots, low DNA area, high DNA area, low DNA amount, high DNA amount, correlation, homogeneity, entropy, fractal dimension, DNA index, OD maximum, and medium DNA amount.

Annuity Stage	Window Opens	Payment Deadline	Surcharge Period Until	Estimated Fee	Status
3.5 Years (Year 4)	2003-02-14	2003-08-16	2004-02-15	$2000.00	Window Closed / Lapsed
7.5 Years (Year 8)	2007-02-14	2007-08-16	2008-02-15	$3760.00	Window Closed / Lapsed
11.5 Years (Year 12)	2011-02-14	2011-08-16	2012-02-15	$7700.00	Window Closed / Lapsed

System and method for automatically detecting malignant cells and cells having malignancy-associated changes

Abstract

Inventors

Bibliographic Details

Claims Analysis

References Cited (Backward Citations)

USPTO Annuity & Maintenance Windows

Assigned Categories & Tags

Assign Category Tag