Generating image-based diagnostic tests by optimizing image analysis and data mining of co-registered images
First Claim
1. A method comprising:
- (a) generating first objects that are linked to pixels of a digital image of a slice of tissue of a patient, wherein a first rule set defines which pixels are linked to the first objects;
(b) using a second rule set to generate a first numerical image layer by measuring a first characteristic of the first objects, wherein each pixel of the first numerical image layer has a value proportional to the first characteristic of the first object linked to that pixel;
(c) generating a heat map by downscaling the first numerical image layer;
(d) generating second objects from the heat map, wherein a third rule set defines which pixels of the heat map are linked to each of the second objects;
(e) using a fourth rule set to generate numerical data by measuring a second characteristic of the second objects detected in the heat map;
(f) determining how well the numerical data correlates with clinical data for the patient;
(g) improving how well the numerical data correlates with the clinical data for the patient by modifying the first rule set, the second rule set, the third rule set and the fourth rule set and then repeating (b) through (e); and
(h) defining an image-based diagnostic test based on the first rule set, the second rule set, the third rule set and the fourth rule set.
1 Assignment
0 Petitions
Accused Products
Abstract
A method for generating an image-based test improves diagnostic accuracy by iteratively modifying rule sets governing image and data analysis of coregistered image tiles. Digital images of stained tissue slices are divided into tiles, and tiles from different images are coregistered. First image objects are linked to selected pixels of the tiles. First numerical data is generated by measuring the first objects. Each pixel of a heat map aggregates first numerical data from coregistered tiles. Second objects are linked to selected pixels of the heat map. Measuring the second objects generates second numerical data. The method improves how well second numerical data correlates with clinical data of the patient whose tissue is analyzed by modifying the rule sets used to generate the first and second objects and the first and second numerical data. The test is defined by those rule sets that produce the best correlation with the clinical data.
-
Citations
20 Claims
-
1. A method comprising:
-
(a) generating first objects that are linked to pixels of a digital image of a slice of tissue of a patient, wherein a first rule set defines which pixels are linked to the first objects; (b) using a second rule set to generate a first numerical image layer by measuring a first characteristic of the first objects, wherein each pixel of the first numerical image layer has a value proportional to the first characteristic of the first object linked to that pixel; (c) generating a heat map by downscaling the first numerical image layer; (d) generating second objects from the heat map, wherein a third rule set defines which pixels of the heat map are linked to each of the second objects; (e) using a fourth rule set to generate numerical data by measuring a second characteristic of the second objects detected in the heat map; (f) determining how well the numerical data correlates with clinical data for the patient; (g) improving how well the numerical data correlates with the clinical data for the patient by modifying the first rule set, the second rule set, the third rule set and the fourth rule set and then repeating (b) through (e); and (h) defining an image-based diagnostic test based on the first rule set, the second rule set, the third rule set and the fourth rule set. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12)
-
-
13. A method comprising:
-
(a) generating first objects that are linked to pixels of a higher resolution digital image of a tissue slice of a patient, wherein a first rule set defines which higher resolution pixels are linked to which of the first objects; (b) using a second rule set to generate a lower resolution digital image comprising heat map pixels, wherein a subset of the heat map pixels have locations on the lower resolution digital image corresponding to where an associated first object is located on the higher resolution digital image, and wherein each of the heat map pixels in the subset has a value corresponding to a first characteristic of the first objects; (c) generating second objects from the lower resolution digital image, wherein a third rule set defines which heat map pixels are linked to each of the second objects; (d) using a fourth rule set to generate numerical data by measuring a second characteristic of the second objects detected in the lower resolution digital image; (e) determining how well the numerical data correlates with clinical data for the patient; (f) improving how well the numerical data correlates with the clinical data for the patient by modifying the second rule set, the third rule set and the fourth rule set and then repeating (b) through (e); and (g) defining an image-based diagnostic test based on the second rule set, the third rule set and the fourth rule set. - View Dependent Claims (14, 15, 16, 17, 18, 19)
-
-
20. A method comprising:
-
(a) generating objects that are linked to pixels of a higher resolution digital image of a tissue slice of a patient and that a classified as belonging to a first class, wherein a first rule set defines which higher resolution pixels are linked to which of the objects and which objects are classified as belonging to the first class; (b) using a second rule set to generate a lower resolution digital image comprising heat map pixels, wherein each heat map pixel corresponds to a region on the higher resolution digital image, and wherein each heat map pixel has a value corresponding to how many objects belong to the first class within the region; (c) generating second objects from the lower resolution digital image, wherein a third rule set defines which heat map pixels are linked to each of the second objects; (d) using a fourth rule set to generate numerical data by measuring a second characteristic of the second objects detected in the lower resolution digital image; (e) determining how well the numerical data correlates with clinical data for the patient; (f) improving how well the numerical data correlates with the clinical data for the patient by modifying the second rule set, the third rule set and the fourth rule set and then repeating (b) through (e); and (g) defining an image-based diagnostic test based on the second rule set, the third rule set and the fourth rule set.
-
Specification