Extracting shape information contained in cell images
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
G06K-009/00
G06K-009/34
출원번호
US-0165914
(2005-06-24)
등록번호
US-7269278
(2007-09-11)
발명자
/ 주소
Cong,Ge
Vaisberg,Eugeni A.
출원인 / 주소
Cytokinetics, Inc.
대리인 / 주소
Beyer Weaver LLP
인용정보
피인용 횟수 :
3인용 특허 :
48
초록▼
Methods and apparatus are provided for the analysis of images of cells and extraction biologically-significant shape-related features from the cell images. The extracted features may be correlated with particular conditions induced by biologically-active agents with which cells have been treated, th
Methods and apparatus are provided for the analysis of images of cells and extraction biologically-significant shape-related features from the cell images. The extracted features may be correlated with particular conditions induced by biologically-active agents with which cells have been treated, thereby enabling the automated analysis of cells based on cell shape parameters. In particular, the invention provides methods for segmentation of cells in an image using a combination of a reference component image data and cell shape-indicative marker image data in a watershed technique. Further, the invention provides a skeletonization and skeleton analysis technique for extracting biologically-relevant features from cell shapes.
대표청구항▼
What is claimed is: 1. A method of identifying boundaries of biological cells, the method comprising: receiving a first image of a field of cells in which a reference cell component of the cells is identified by a reference cell component marker image parameter, wherein the reference cell component
What is claimed is: 1. A method of identifying boundaries of biological cells, the method comprising: receiving a first image of a field of cells in which a reference cell component of the cells is identified by a reference cell component marker image parameter, wherein the reference cell component is selected from the nucleus, centrosome, a chromosome and the Golgi complex; receiving a second image of the field of the cells in which a shape-indicative marker of the cells is identified by a cell shape-indicative marker image parameter; thresholding the cell shape-indicative marker in the second image to generate a digital representation of the second image comprising a cell shape-indicative marker portion and a background portion; segmenting the reference cell component in the first image to generate a digital representation of the first image (reference cell component mask); conceptually registering the reference cell component mask with the digital representation of the second image; and applying a watershed algorithm to data provided by the registered reference component mask and digital representation of the second image to segment the cells in the field such that individual cell boundaries for the cells in the field are identified. 2. The method of claim 1, wherein the cell shape-indicative marker is at least one of a cytoskeletal, a cytoplasmic, and a plasma membrane marker. 3. The method of claim 1, wherein said reference cell component is nucleus and said cell shape-indicative marker is a tubulin marker. 4. The method of claim 1, wherein said thresholding of the cell shape-indicative marker comprises: converting the second image to a digital representation of the second image, wherein a pixel having an image parameter intensity greater than a threshold intensity, ITH, is recognized as the cell shape-indicative marker and is assigned one of 0 and non-zero, and wherein a pixel having an image parameter intensity less than the threshold intensity, ITH, is recognized as background and is assigned the other of 0 and non-zero; wherein said threshold is calculated according to a method comprising, generating a histogram of number of pixels versus image parameter intensity, assigning an intensity of the greatest number of pixels, IMAX, as background intensity, determining a standard deviation of a normal distribution of the background intensity, ISTD, and description="In-line Formulae" end="lead"assigning a value to ITH=IMAX+c*ISTD. description="In-line Formulae" end="tail" 5. The method of claim 4, wherein 0.7>c>0.9. 6. The method of claim 1, wherein the application of the watershed algorithm to the data provided by the conceptually registered reference cell component mask and digital representation of the second image uses the original cell image and two different types of seeds. 7. The method of claim 6, wherein the seeds are the reference cell component portion of the reference component mask and the background portion of the digital representation of the second image. 8. The method of claim 1 wherein at least two of the cells in the field of cells overlap and/or abut each other. 9. A method of identifying boundaries of biological cells, the method comprising: receiving a first image of a field of cells in which a reference cell component of the cells is identified by a reference cell component marker image parameter, wherein the reference cell component is selected from the nucleus, centrosome, a chromosome and the Golgi complex; receiving a second image of the field of cells in which at least one of a cell shape-indicative marker of the cells is identified by a cell shape-indicative marker image parameter; thresholding the cell shape-indicative marker image parameter in the second image to generate a digital representation of the second image comprising a cell shape-indicative marker portion and a background portion; and identifying boundaries of individual cells by applying a watershed algorithm to the second image using the reference cell component marker image parameter and the background portion of the digital representation of the second image as seeds. 10. The method of claim 9, wherein the cell shape-indicative marker is at least one of a cytoskeletal, a cytoplasmic, and a plasma membrane marker. 11. The method of claim 9, wherein said reference cell component is nucleus and said cell shape-indicative marker is a tubulin marker. 12. A method of extracting biologically-significant shape-related information from a field of one or more cells, comprising: (a) providing a segmented image of the field of one or more segmented cells, the boundaries of said one or more segmented cells having been ascertained by the segmentation; for each of one or more of the cells in the segmented cell image, (b) selecting two endpoints defining two parts of the boundary of at least one of said one or more cells; (c) for each part of said cell boundary, computing the distance from each point on the part of the cell boundary to a line between said endpoints; (d) determining a point dMAX on the portion of the boundary, said point dMAX being maximally distant from the line; (e) comparing the distance from the point dMAX to a predetermined threshold distance value, dTH; (f) where dMAX is greater than dTH, discarding the line between the endpoints, using point dMAX as a new endpoint together with one of the original endpoints to separate the part into two new parts, and repeating (c) and following; (g) where dMAX is less than dTH, using the line as a side of a polygon approximating the cell shape until a polygon approximating the shape of the cell is complete; and (h) skeletonizing and computing at least one of end points and nodes for the polygon approximation of the cell. 13. A method of correlating a cell's shape with a biological condition of the cell, comprising: (a) providing a plurality of segmented images of fields of one or more segmented cells, at least one of said fields having been treated with a biologically active agent and at least one of said fields being a control and having not been treated with the biologically active agent, the boundaries of said one or more segmented cells having been ascertained by the segmentation; for each of one or more of the cells in the plurality of segmented cell images, (b) selecting two endpoints defining two parts of the boundary of said cell; (c) for each part of said cell boundary, computing the distance from each point on the part of the cell boundary to a line between said endpoints; (d) determining a point dMAX on the portion of the boundary, said point dMAX being maximally distant from the line; (e) comparing the distance from the point dMAX to a predetermined threshold distance value, dTH; (f) where dMAX is greater than dTH, discarding the line between the endpoints, using point dMAX as a new endpoint together with one of the original endpoints to separate the part into two new parts, and repeating (c) and following; (g) where dMAX is less than dTH, using the line as a side of a polygon approximating the cell shape until a polygon approximating the shape of the cell is complete; (h) skeletonizing and computing at least one of end points and nodes for the polygon approximation of the cell; and (i) comparing the computations of the at least one of end points and nodes for the polygon approximation of the cell to identify significant shape differences between the treated and control fields of one or more cells. 14. The method of claim 13, wherein said segmentation comprises, for each field: receiving a first image of a field of one or more cells in which a reference cell component of the one or more cells is identified by a reference cell component marker image parameter; receiving a second image of the field of one or more cells in which at least one of a cell shape-indicative marker of the one or more cells is identified by a cell shape-indicative marker image parameter; and processing the first image in conjunction with the second image such that individual cell boundaries for the one or more cells in the field are identified. 15. The method of claim 13, wherein said computing of at least one of end points and nodes for the polygon approximation of the cell comprises quantifying the at least one of end points and nodes for the polygon approximation of the cell. 16. A method of identifying boundaries of biological cells, the method comprising: receiving a first image of a field of cells in which a reference cell component of the cells is identified by a reference cell component marker image parameter; receiving a second image of the field of the cells in which a shape-indicative marker of the cells is identified by a cell shape-indicative marker image parameter, wherein the cell shape-indicative marker is at least one of a cytoskeletal, a cytoplasmic, and a plasma membrane marker; thresholding the cell shape-indicative marker in the second image to generate a digital representation of the second image comprising a cell shape-indicative marker portion and a background portion; segmenting the reference cell component in the first image to generate a digital representation of the first image (reference cell component mask); conceptually registering the reference cell component mask with the digital representation of the second image; and applying a watershed algorithm to data provided by the registered reference component mask and digital representation of the second image to segment the cells in the field such that individual cell boundaries for the cells in the field are identified. 17. The method of claim 16, wherein the reference cell component is selected from the nucleus, centrosome, a chromosome and the Golgi complex. 18. The method of claim 16, wherein said reference cell component is nucleus and said cell shape-indicative marker is a tubulin marker. 19. The method of claim 16, wherein said thresholding of the cell shape-indicative marker comprises: converting the second image to a digital representation of the second image, wherein a pixel having an image parameter intensity greater than a threshold intensity, ITH, is recognized as the cell shape-indicative marker and is assigned one of 0 and non-zero, and wherein a pixel having an image parameter intensity less than the threshold intensity, ITH, is recognized as background and is assigned the other of 0 and non-zero; wherein said threshold is calculated according to a method comprising, generating a histogram of number of pixels versus image parameter intensity, assigning an intensity of the greatest number of pixels, IMAX, as background intensity, determining a standard deviation of a normal distribution of the background intensity, ISTD, and description="In-line Formulae" end="lead"assigning a value to ITH=IMAX+c*ISTD. description="In-line Formulae" end="tail" 20. The method of claim 19, wherein 0.7>c>0.9. 21. The method of claim 16, wherein the application of the watershed algorithm to the data provided by the conceptually registered reference cell component mask and digital representation of the second image uses the original cell image and two different types of seeds. 22. The method of claim 21, wherein the seeds are the reference cell component portion of the reference component mask and the background portion of the digital representation of the second image. 23. The method of claim 16 wherein at least two of the cells in the field of cells overlap and/or abut each other. 24. A method of identifying boundaries of biological cells, the method comprising: receiving a first image of a field of cells in which a reference cell component of the cells is identified by a reference cell component marker image parameter; receiving a second image of the field of cells in which at least one of a cell shape-indicative marker of the cells is identified by a cell shape-indicative marker image parameter, wherein the cell shape-indicative marker is at least one of a cytoskeletal, a cytoplasmic, and a plasma membrane marker; thresholding the cell shape-indicative marker image parameter in the second image to generate a digital representation of the second image comprising a cell shape-indicative marker portion and a background portion; and identifying boundaries of individual cells by applying a watershed algorithm to the second image using the reference cell component marker image parameter and the background portion of the digital representation of the second image as seeds. 25. The method of claim 24, wherein the reference cell component is selected from the nucleus, centrosome, a chromosome and the Golgi complex. 26. The method of claim 24 wherein said reference cell component is nucleus and said cell shape-indicative marker is a tubulin marker.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (48)
Price Jeffrey H. ; Bravo-Zanoguera Miguel, Analog circuit for an autofocus microscope system.
Rutenberg Mark R. (Monsey NY) Hall Thomas L. (Pasadena CA) Domes Robert (Newbury Park CA) Chaban Richard (Ventura CA) Dulak Thomas (Newhall CA) Herriman James M. (New York NY) Porzio John (New York N, Automated cytological specimen classification system and method.
Ravkin Ilya, Identification of objects of interest using multiple illumination schemes and finding overlap of features in corresponding multiple images.
Sutherland Ranald M. (Geneva CHX) Dhne Claus (Onex CHX) Revillet Georges (Onex CHX), Method for optically ascertaining parameters of species in a liquid analyte.
Singhvi Rahul ; Kumar Amit ; Whitesides George M. ; Ingber Donald E. ; Lopez Gabriel P. ; Wang Daniel I. C. ; Stephanopoulos Gregory N., Method of formation of microstamped patterns on plates for adhesion of cells and other biological materials, devices an.
Hozier John (8950 S. Tropical Trail Merritt Island FL 32952), Micro-libraries and methods of making and manipulating them methods for generating and analyzing micro-libraries.
Weaver James C. (Sudbury MA) Bliss Jonathan G. (Somerville MA) Williams Gregory B. (Tewksbury MA) Powell Kevin T. (Boston MA) Harrison Gail I. (Watertown MA), Process for rapidly enumerating viable entities.
Jain Ramesh ; Horowitz Bradley ; Fuller Charles E. ; Gupta Amarnath ; Bach Jeffrey R. ; Shu Chiao-fe, Similarity engine for content-based retrieval of images.
Odeyale Charles O. (Severn MD) Hook Gregory R. (Silver Spring MD), System and method for quantifying macrophage phagocytosis by computer image analysis.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.