System and methods for mapping fluorescent images into a bright field color space
View Patent ↗A method for generating a brightfield type image, which resembles a brightfield staining protocol of a biological sample, using fluorescent images is provided. The steps comprise acquiring two or more fluorescent images of a fixed area on a biological sample, mapping said fluorescent image into a brightfield color space, and generating a bright field image. Also provided is an image analysis system for generating a brightfield type image of a biological sample using fluorescent images.
1. A method for generating a brightfield type image, that resembles a brightfield staining protocol of a biological sample, using fluorescent images comprising the steps of:
acquiring, by a digital imaging device, two or more fluorescent images of a fixed area on a biological sample;
mapping said fluorescent images into a brightfield color space; and
generating a bright field type image;
the mapping step comprises:
acquiring a bright field image of the fixed area of the biological sample;
analyzing the image data of the bright field image and the two or more fluorescent images utilizing, at least in part, featured based information or pixel intensity data information to generate mapping parameters to transform the two or more fluorescent images into a bright field color space; and
applying said mapping parameters to the two or more fluorescent images;
the generating mapping parameters step comprises:
using a linear estimation model;
the linear estimation model is defined as:
A
^
=
arg
min
A
(
HE
-
A
·
FL
)
2
where  is the estimated mapping parameters, HE represents intensity values in the color channels of the Hematoxylin & Eosin (H&E) image, and FL is the intensity values of at least two color channels of the fluorescent images.
2. The method of claim 1 wherein the brightfield type image corresponds to an Hematoxylin & Eosin (H&E) type image having a red, green, and blue three channel color space.
3. The method of claim 1 wherein at least one image of the two or more fluorescent images is of autofluorescence.
4. The method of claim 1 wherein the acquiring a bright field image step comprises the steps of sequentially staining the biological sample with hematoxylin and eosin to generate an Hematoxylin & Eosin (H&E) type image.
5. The method of claim 1 wherein the feature based information comprises one or more features selected from a group consisting of nuclei, epithelia, and stroma.
6. The method of claim 1 further comprising the step of applying the mapping parameters to two or more fluorescent images of a second fixed area wherein the second fixed area is from the same or a different biological sample.
7. The method of claim 1 further comprising the step of pathological diagnostics using said brightfield type image.
8. The method of claim 7 wherein the pathological diagnostics is for cancer.
9. The method of claim 1 further comprising the step of quantitative analysis using said brightfield type image.
10. The method of claim 9 wherein the step of quantitative analysis comprises identifying molecular pathways as a function of one or more morphological structures selected from a group consisting of nuclei, epithelia, and stroma.
11. An image analysis system for generating a brightfield type image, that resembles a brightfield staining protocol of a biological sample, using fluorescent images comprising:
a digital imaging device to acquire two or more fluorescent images of a fixed area on a biological sample; and
a processing device to apply mapping parameters to transform the two or more fluorescent images into a brightfield type image;
the processing device is to calculate the mapping parameters by:
acquiring a bright field image of the fixed area of the biological sample;
analyzing at least in part, feature based information or pixel intensity data information of the bright field image and the two or more fluorescent images to transform the two or more fluorescent images into a bright field color space;
the analyzing step comprises a linear estimation model;
the linear estimation model is defined as:
A
^
=
arg
min
A
(
HE
-
A
·
FL
)
2
where  is the estimated mapping parameters, HE represents intensity values in the color channels of the Hematoxylin & Eosin (H&E) image, and FL is the intensity values of at least two color channels of the fluorescent images.
12. The system of claim 11 wherein the processing device is further configured to store mapping parameters from one or more previously analyzed biological samples.