SYSTEMS AND METHODS FOR PERFORMING CORRELATION ANALYSIS ON CLINICAL OUTCOME AND CHARACTERISTICS OF BIOLOGICAL TISSUE

• US 20130290006A1
• Filed: 04/30/2012
• Published: 10/31/2013
• Est. Priority Date: 04/30/2012
• Status: Active Grant

• Current Assignee: Leica Microsystems CMS GmbH (Danaher Corporation)
• Original Assignee: General Electric Company
• Inventors: Kamath, Vidya Pundalik, McCulloch, Colin Craig, Sarachan, Brion Daryl, Rothney, Megan, Santamaria-Pang, Alberto
• Granted Patent: US 8,737,709 B2
• Time in Patent Office: Days
• US Class Current: 705/2
• CPC Class Codes: G16H 30/40 for processing medical images, e.g. editing
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

Embodiments disclosed herein include methods, systems, and devices for determining a positive or negative correlation between a clinical outcome and one or more features of biological tissue. An exemplary user interface enables a user to select a clinical outcome and one or more aspects of a displayed field-of-view of biological tissue. Exemplary embodiments may automatically perform correlation analysis between the selected clinical outcome and one or more features characteristic of the user-selected aspects of the field-of-view of biological tissue. For example, exemplary embodiments may automatically perform correlation analysis between the selected clinical outcome and one or more features characteristic of one or more biological units in the biological tissue.

Claims

1. A computer-implemented method for correlating one or more features of biological tissue with a clinical outcome, the method comprising:

rendering a graphical user interface on a visual display device;

rendering, on the graphical user interface, a field of view selection component allowing a user to select a field of view from a data set of a cohort comprising tissue profile data including multiplexed biomarker images capturing expression of a plurality of biomarkers in a plurality of fields of view of biological tissue;

in response to user input selecting a first biomarker and a first field of view corresponding to a biological tissue at the field of view selection component, rendering, on the graphical user interface, a first image of the selected first field of view corresponding to the biological tissue, the first image representing expression levels of the selected first biomarker and including representations of one or more biological units in the biological tissue;

receiving, at the graphical user interface, a selection of a clinical outcome and a selection of a first set of one or more of the biological units in the biological tissue;

selectively displaying or highlighting the selected first set of biological units represented on the first image; and

automatically determining and displaying a result of a correlation analysis performed between one or more features of the selected first set of biological units in the cohort and the selected clinical outcome.

2. The method of claim 1, further comprising:

rendering, on the graphical user interface, a clinical outcome selection component allowing the user to define a clinical outcome selection on the graphical user interface; and

receiving the clinical outcome selection from a user defining a clinical outcome selection directly on the graphical user interface.

3. The method of claim 1, further comprising:

rendering, on the graphical user interface, a biological unit selection component allowing the user to select one or more biological units directly on the first image rendered on the graphical user interface; and

receiving the selection of the first set of the biological units from a user selecting the biological units directly on the first image rendered on the graphical user interface.

4. The method of claim 3, further comprising:

before displaying the result of the correlation analysis, applying a supervised learning algorithm to expand the selection of the first set of the biological units with biological units in the cohort having one or more similar features.

5. The method of claim 1, wherein the selection of the first set of the biological units is made automatically by a clustering algorithm implemented on a computing device that clusters the biological units.

6. The method of claim 5, wherein the clustering algorithm selects the first set of the biological units based on expression levels of one or more biomarkers in the biological units.

7. The method of claim 5, wherein the clustering algorithm selects the first set of the biological units based on expression levels of one or more biomarkers in the biological units and one or more morphological features of the biological units.

8. The method of claim 1, wherein the biological unit is a cell.

9. The method of claim 1, wherein the biological unit is a sub-cellular component of a cell.

10. The method of claim 1, wherein the clinical outcome is one of a positive or negative diagnosis of a disease condition of the biological tissue, a disease prognosis, a prediction of drug response, or a clinically-relevant group.

11. The method of claim 1, wherein the automatic determining of the result of the correlation analysis comprises:

correlating expression levels of one or more biomarkers at the biological units with the clinical outcome.

12. The method of claim 1, wherein the automatic determining of the result of the correlation analysis comprises:

correlating one or more morphological features of the biological units with the clinical outcome.

13. The method of claim 1, wherein the automatic determining of the result of the correlation analysis comprises:

correlating the clinical outcome with the expression levels of one or more biomarkers at the biological units and the one or more morphological features of the biological units.

14. The method of claim 1, further comprising:

determining, based on the result of the correlation analysis, that the one or more features of the selected first set of biological units are positively correlated or negatively correlated with the selected clinical outcome; and

generating a predictive model for the clinical outcome based on the result of the correlation analysis.

15. The method of claim 1, wherein more than six images of the selected field of view corresponding to the biological tissue are rendered in an overlaid manner on the graphical user interface, each image representing expression levels of a biomarker in a different color.

16. The method of claim 1, wherein the rendering of the first image comprises:

generating a set of one or more segments by running an image segmentation algorithm implemented on a computing device on the image data of the selected field of view corresponding to the biological tissue; and

displaying the set of segments on the first image to delineate the one or more biological units.

17. The method of claim 1, further comprising:

sending instructions to store, on a storage device, the result of the correlation analysis between the one or more features of the selected first set of biological units and the selected clinical outcome.

18. A computer-implemented method for correlating one or more features of biological tissue with a clinical outcome, the method comprising:

rendering a graphical user interface on a visual display device;

rendering, on the graphical user interface, a field of view selection component allowing a user to select a field of view from a data set of a cohort comprising tissue profile data including multiplexed biomarker images capturing expression of a plurality of biomarkers in a plurality of fields of view of biological tissue;

in response to user input selecting a first biomarker and a first field of view corresponding to a biological tissue at the field of view selection component, rendering, on the graphical user interface, a first image of the selected first field of view corresponding to the biological tissue, the first image representing expression levels of the selected first biomarker and including representations of one or more biological units in the biological tissue;

receiving, at the graphical user interface, a selection of a first set of one or more of the biological units in the biological tissue;

selectively displaying or highlighting the selected first set of biological units represented on the first image; and

automatically determining and displaying a result of a correlation analysis performed between a clinical outcome and one or more features in the data set of the cohort that are characteristic of the selected first set of biological units.

19. The method of claim 18, wherein the one or more features in the data set of the cohort that are characteristic of the selected first set of biological units are determined by applying a supervised learning algorithm to the selected first set of the biological units and the data set of the cohort.

20. A computer system for correlating one or more features of biological tissue with a clinical outcome, the system comprising:

a visual display device;

a data storage device for storing a data set of a cohort comprising tissue profile data including multiplexed biomarker images capturing expression of a plurality of biomarkers in a plurality of fields of view of biological tissue; and

a computer processor programmed to;

render a graphical user interface on the visual display device,

render, on the graphical user interface, a field of view selection component allowing a user to select a field of view from the data set stored on the data storage device,

in response to user input selecting a first biomarker and a first field of view corresponding to a biological tissue at the field of view selection component, render, on the graphical user interface, a first image of the selected first field of view corresponding to the biological tissue, the first image representing expression levels of the selected first biomarker and including representations of one or more biological units in the biological tissue,

receive, at the graphical user interface, a selection of a clinical outcome and a selection of a first set of one or more of the biological units in the biological tissue,

selectively display or highlight the selected first set of biological units represented on the first image, and

automatically determine and display a result of a correlation analysis performed between one or more features of the selected first set of biological units in the cohort and the selected clinical outcome.

21. A computer system for correlating one or more features of biological tissue with a clinical outcome, the system comprising:

a visual display device;

a data storage device for storing a data set of a cohort comprising tissue profile data including multiplexed biomarker images capturing expression of a plurality of biomarkers in a plurality of fields of view of biological tissue; and

a computer processor programmed to;

render a graphical user interface on the visual display device,

render, on the graphical user interface, a field of view selection component allowing a user to select a field of view from the data set stored on the data storage device,

in response to user input selecting a first biomarker and a first field of view corresponding to a biological tissue at the field of view selection component, render, on the graphical user interface, a first image of the selected first field of view corresponding to the biological tissue, the first image representing expression levels of the selected first biomarker and including representations of one or more biological units in the biological tissue,

receive, at the graphical user interface, a selection of a first set of one or more of the biological units in the biological tissue,

selectively display or highlight the selected first set of biological units represented on the first image, and

automatically determine and display a result of a correlation analysis performed between a clinical outcome and one or more features in the data set of the cohort that are characteristic of the selected first set of biological units.