Method for analyzing droplets based on images, computer device and storage medium

Patent metadata

Field	Value
Publication number	US-12340542-B2
Application number	US-202017799488-A
Country	US
Kind code	B2
Filing date	Feb 14, 2020
Priority date	Feb 14, 2020
Publication date	Jun 24, 2025
Grant date	Jun 24, 2025

Abstract

A method for analyzing droplets based on images: obtaining a working dye channel image and a reference dye channel image corresponding to a droplet system (S 1 ); separately performing a preset image processing operation on the working dye channel image and the reference dye channel image (S 2 ); using the working dye channel image or the reference dye channel image as a target image (S 3 ), and determining a region in which each droplet is located from within the target image (S 4 ); and on the basis of the determined region in which each droplet is located, analyzing the droplet system (S 5 ). The method can effectively improve the accuracy of droplet analysis. Further provided are a computer device and a storage medium.

First claim

What is claimed is:

1. A method for analyzing droplets based on images, comprises:
acquiring a working dye channel image and a reference dye channel image corresponding to a droplet system;
separately performing a preset image processing operation on the working dye channel image and the reference dye channel image;
taking the working dye channel image or the reference dye channel image as a target image;
determining a region where each droplet is located from the target image; and
analyzing the droplet system based on the determined region where each droplet is located.

2. The method for analyzing droplets based on images according to claim 1 , wherein the acquiring the working dye channel image and the reference dye channel image corresponding to the droplet system comprises: exciting a working dye and a reference dye in the droplet system separately under a condition of a dual/multi-channel wavelength, obtaining the working dye channel image by capturing a fluorescence image of a channel corresponding to the working dye, and obtaining the reference dye channel image by capturing a fluorescence image of a channel corresponding to the reference dye.

3. The method for analyzing droplets based on images according to claim 1 , wherein the separately performing the preset image processing operation on the working dye channel image and the reference dye channel image comprises:
separately performing a correction operation for uneven background brightness on the working dye channel image and the reference dye channel image;
separately performing an image enhancement processing on the reference dye channel image and the working dye channel image by using a preset image enhancement algorithm; and
separately performing a noise reduction processing on the working dye channel image and the reference dye channel image by using a preset noise reduction algorithm.

4. The method for analyzing droplets based on images according to claim 1 , wherein the determining the region where each droplet is located from the target image comprises:
performing a local contrast enhancement processing on the target image;
obtaining a grayscale image by performing a grayscale conversion on the target image, and obtaining a binarized target image by binarizing the grayscale image;
dividing the binarized target image into several discrete image regions, and each image region in the several discrete image regions being corresponding to a single droplet, thereby determining the region where each droplet is located.

5. The method for analyzing droplets based on images according to claim 1 , wherein before the determining of the region where each droplet is located from the target image, the method further comprises:
determining one of the working dye channel image and the reference dye channel image that is not taken as the target image as a non-target image, taking the target image as a fixed benchmark, and performing a registration operation on the non-target image.

6. The method for analyzing droplets based on images according to claim 5 , wherein the taking of the target image as the fixed benchmark, and the performing of the registration operation on the non-target image comprise:
taking the target image as the fixed benchmark; obtaining a first transformation matrix by performing a first image registration operation on the non-target image; and obtaining a first registration image based on the first transformation matrix and the non-target image;
taking the target image as the fixed benchmark; obtaining a second transformation matrix by performing a second image registration operation on the first registration image; obtaining a second registration image based on the second transformation matrix and the first registration image; and determining the second registration image as a registered non-target image.

7. The method for analyzing droplets based on images according to claim 1 , wherein the analyzing the droplet system based on the determined region where each droplet is located comprises:
calculating an average light intensity (I Abs ) of each droplet on the working dye channel image according to the region where each droplet is located, and calculating an average light intensity (I ref ) of each droplet on the reference dye channel image according to the region where each droplet is located;
calculating the relative light intensity (I rela ) of each droplet based on the average light intensity (I Abs ) of each droplet on the working dye channel and the average light intensity (I Ref ) of each droplet on the reference dye channel image, and I Rela =I Abs /I Ref .

8. A computer device, characterized in that the computer device comprises a storage device and a processor, the storage device storing at least one computer readable instruction, which when executed by the processor, cause the processor to: acquire a working dye channel image and a reference dye channel image corresponding to a droplet system;
separately perform a preset image processing operation on the working dye channel image and the reference dye channel image;
take the working dye channel image or the reference dye channel image as a target image;
determine a region where each droplet is located from the target image; and
analyze the droplet system based on the determined region where each droplet is located.

9. The computer device according to claim 8 , wherein the processor is further caused to: excite a working dye and a reference dye in the droplet system separately under a condition of a dual/multi-channel wavelength, obtain the working dye channel image by capturing a fluorescence image of a channel corresponding to the working dye, and obtain the reference dye channel image by capturing a fluorescence image of a channel corresponding to the reference dye.

10. The computer device according to claim 8 , wherein the processor is further caused to:
separately perform a correction operation for uneven background brightness on the working dye channel image and the reference dye channel image;
separately perform an image enhancement processing on the reference dye channel image and the working dye channel image by using a preset image enhancement algorithm; and
separately perform a noise reduction processing on the working dye channel image and the reference dye channel image by using a preset noise reduction algorithm.

11. The computer device according to claim 8 , wherein the processor is further caused to: perform a local contrast enhancement processing on the target image;
obtain a grayscale image by performing a grayscale conversion on the target image, and obtain a binarized target image by binarizing the grayscale image;
divide the binarized target image into several discrete image regions, and each image region in the several discrete image regions being corresponding to a single droplet, thereby the region where each droplet is located is determined.

12. The computer device according to claim 8 , wherein the processor is further caused to: determine one of the working dye channel image and the reference dye channel image that is not taken as the target image as a non-target image, take the target image as a fixed benchmark, and perform a registration operation on the non-target image.

13. The computer device according to claim 12 , wherein the taking of the target image as the fixed benchmark, and the performing of the registration operation on the non-target image, comprise:
taking the target image as the fixed benchmark; obtaining a first transformation matrix by performing a first image registration operation on the non-target image; and obtaining a first registration image base

Assignees

• Mgi Tech Co Ltd

Inventors

• Xia Yun
• Xi Yang
• Zhao Xia
• Cui xingye
• Shen Hanjie
• Chen Fang
• Jiang Hui

Classifications

• G06T5/94

  based on local image properties, e.g. for local contrast enhancement · CPC title

• G06T5/70

  Denoising; Smoothing · CPC title

• G06T2207/10064

  Fluorescence image · CPC title

• G06T7/00

  Image analysis · CPC title

• G01N2021/6439

  with indicators, stains, dyes, tags, labels, marks · CPC title