# US 12,111,244 B2 · Method for calculating a density of stem cells in a cell image, electronic device, and storage medium > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,111,244 B2 - Document ID: US12111244B2 - Status: Granted - Kind: B2 - Application number: 17/523,989 - Filed: 2021-11-11 - Published: 2022-06-09 - Granted: 2024-10-08 - Earliest claimed: 2020-12-03 - Assignees: Hon Hai Precision Industry Co., Ltd. - Applicants: Hon Hai Precision Industry Co., Ltd. - Inventors: Wan-Jhen Lee; Chin-Pin Kuo; Chih-Te Lu - Web page: https://patentgrep.com/patent/US12111244B2 ## CPC - G01N 15/1433: using image recognition - G01N 15/1429: Signal processing - G01N 33/4833: {of solid biological material, e.g. tissue samples, cell cultures (tissue in vivo A61B5/00; cell suspensions G01N33/48735)} - G06T 7/0012: {Biomedical image inspection} - G06V 10/774: Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting - G06V 20/698: {Matching; Classification} - G01N 2015/1006: {for cytology} - G06T 2207/20081: Training; Learning - G06T 2207/30024: Cell structures in vitro; Tissue sections in vitro - G06T 2207/30242: Counting objects in image - G01N: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES (measuring or testing apparatus or processes other than immunoassay, involving enzymes or microorganisms C12M, C12Q) - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract A method for calculating a density of stem cells in a cell image and an electronic device are provided. A plurality of preset ratios and a plurality of density calculation models can be used to perform hierarchical density calculations on the cell image. Starting from the largest preset ratio (the first preset ratio) reduction of the cell image to no reduction, the density calculation is performed on the cell image using a model starting with a highest density calculation (the first density calculation model) to a model with the smallest density calculation (the third density calculation model), which can quickly detect densities of various stem cells. Using different preset ratios and corresponding density calculation models for calculation, it is not necessary to calculate the number of stem cells to obtain the density of stem cells, which improves a calculation efficiency of the density of stem cells. ## Claims ### Claim 1 A method for calculating a density of stem cells in a cell image, the method comprising: acquiring a cell image; reducing an image size of the cell image according to a preset first ratio and obtaining a first reduced image; calculating a density of stem cells in the first reduced image by using a pre-trained first density calculation model; and when the density of stem cells in the first reduced image is less than a preset first density, calculating at least one density of stem cells in the cell image according to a plurality of preset second ratios and a plurality of pre-trained second density calculation models, wherein the plurality of preset second ratios are in one-to-one correspondence with the plurality of pre-trained second density calculation models. ### Claim 2 (depends on 1) The method for calculating a density of stem cells in a cell image of claim 1, the method further comprising: when the at least one density of stem cells in the cell image is less than a preset second density, calculating a density of stem cells in the cell image by using a pre-trained third density calculation model. ### Claim 3 (depends on 1) The method for calculating a density of stem cells in a cell image of claim 1, wherein calculating at least one density of stem cells in the cell image according to a plurality of preset second ratios and a plurality of pre-trained second density calculation models comprises: obtaining a largest second ratio among the plurality of preset second ratios; reducing the image size of the cell image according to the largest second ratio and obtaining a second reduced image; calculating a density of stem cells in the second reduced image by using a pre-trained second density calculation model corresponding to the largest second ratio; when the density of stem cells in the second reduced image is less than a preset second density corresponding to the largest second ratio, obtaining a second largest second ratio among the plurality of preset second ratios; reducing the image size of the cell image according to the second largest second ratio and obtaining a third reduced image; and calculating a density of stem cells in the third reduced image by using a pre-trained second density calculation model corresponding to the second largest second ratio. ### Claim 4 (depends on 3) The method for calculating a density of stem cells in a cell image of claim 3, the method further comprising: if the density of stem cells in the second reduced image is greater than or equal to the preset second density corresponding to the largest second ratio, determining that the density of stem cells in the cell image is greater than or equal to the preset second density corresponding to the largest second ratio but less than the preset first density. ### Claim 5 (depends on 4) The method for calculating a density of stem cells in a cell image of claim 4, before acquiring the cell image, the method further comprising: acquiring a plurality of first training images; selecting, from the plurality of first training images, images with a density of stem cells greater than or equal to the preset first density as first positive sample images, and images with a density of stem cells less than the preset first density as first negative sample images; reducing image sizes of the first positive sample images and image sizes of the first negative sample images according to the preset first ratio; and training and obtaining the first density calculation model with the reduced first positive sample images and the reduced first negative sample images. ### Claim 6 (depends on 5) The method for calculating a density of stem cells in a cell image of claim 5, before acquiring the cell image, the method further comprising: acquiring a plurality of second training images; for a target second ratio of the plurality of preset second ratios, selecting, from the plurality of second training images, images with a density of stem cells greater than or equal to a preset second density corresponding to the target second ratio as second positive sample images, and images with a density of stem cells less than the preset second density corresponding to the target second ratio as second negative sample images; reducing image sizes of the second positive sample images and image sizes of the second negative sample images according to the target second ratio; and training and obtaining a second density calculation model corresponding to the target second ratio with the reduced second positive sample images and the reduced second negative sample images. ### Claim 7 (depends on 6) The method for calculating a density of stem cells in a cell image of claim 6, before acquiring the cell image, the method further comprising: acquiring a plurality of third training images; selecting, from the plurality of third training images, images with a density of stem cells greater than or equal to a preset third density as third positive sample images, and images with a density of stem cells less than the preset third density as third negative sample images; training and obtaining the third density calculation model with the reduced third positive sample images and the reduced third negative sample images. ### Claim 8 A electronic device comprising a memory and a processor, the memory stores at least one computer-readable instruction, and the processor executes the at least one computer-readable instruction to: acquire a cell image; reduce an image size of the cell image according to a preset first ratio and obtain a first reduced image; calculate a density of stem cells in the first reduced image by using a pre-trained first density calculation model; and when the density of stem cells in the first reduced image is less than a preset first density, calculate at least one density of stem cells in the cell image according to a plurality of preset second ratios and a plurality of pre-trained second density calculation models, wherein the plurality of preset second ratios are in one-to-one correspondence with the plurality of pre-trained second density calculation models. ### Claim 9 (depends on 8) The electronic device of claim 8, wherein the processor further to: when the at least one density of stem cells in the cell image is less than a preset second density, calculate a density of stem cells in the cell image by using a pre-trained third density calculation model. ### Claim 10 (depends on 8) The electronic device of claim 8, wherein calculating at least one density of stem cells in the cell image according to a plurality of preset second ratios and a plurality of pre-trained second density calculation models comprises: obtaining a largest second ratio among the plurality of preset second ratios; reducing the image size of the cell image according to the largest second ratio and obtaining a second reduced image; calculating a density of stem cells in the second reduced image by using a pre-trained second density calculation model corresponding to the largest second ratio; when the density of stem cells in the second reduced image is less than a preset second density corresponding to the largest second ratio, obtaining a second largest second ratio among the plurality of preset second ratios; reducing the image size of the cell image according to the second largest second ratio and obtaining a third reduced image; and calculating a density of stem cells in the third reduced image by using a pre-trained second density calculation model corresponding to the second largest second ratio. ### Claim 11 (depends on 10) The electronic device of claim 10, wherein the processor further to: if the density of stem cells in the second reduced image is greater than or equal to the preset second density corresponding to the largest second ratio, determining that the density of stem cells in the cell image is greater than or equal to the preset second density corresponding to the largest second ratio but less than the preset first density. ### Claim 12 (depends on 11) The electronic device of claim 11, before acquiring the cell image, wherein the processor further to: acquire a plurality of first training images; select, from the plurality of first training images, images with a density of stem cells greater than or equal to the preset first density as first positive sample images, and images with a density of stem cells less than the preset first density as first negative sample images; reduce image sizes of the first positive sample images and image sizes of the first negative sample images according to the preset first ratio; and train and obtain the first density calculation model with the reduced first positive sample images and the reduced first negative sample images. ### Claim 13 (depends on 12) The electronic device of claim 12, before acquiring the cell image, the processor further to: acquire a plurality of second training images; for a target second ratio of the plurality of preset second ratios, select, from the plurality of second training images, images with a density of stem cells greater than or equal to a preset second density corresponding to the target second ratio as second positive sample images, and images with a density of stem cells less than the preset second density corresponding to the target second ratio as second negative sample images; reduce image sizes of the second positive sample images and image sizes of the second negative sample images according to the target second ratio; and train and obtain a second density calculation model corresponding to the target second ratio with the reduced second positive sample images and the reduced second negative sample images. ### Claim 14 (depends on 13) The electronic device of claim 13, before acquiring the cell image, wherein the processor further to: acquire a plurality of third training images; select, from the plurality of third training images, images with a density of stem cells greater than or equal to a preset third density as third positive sample images, and images with a density of stem cells less than the preset third density as third negative sample images; and train and obtain the third density calculation model with the reduced third positive sample images and the reduced third negative sample images. ### Claim 15 A non-transitory storage medium having stored thereon at least one computer-readable instructions that, when the at least one computer-readable instructions are executed by a processor to implement the following method: acquiring a cell image; reducing an image size of the cell image according to a preset first ratio and obtaining a first reduced image; calculating a density of stem cells in the first reduced image by using a pre-trained first density calculation model; and when the density of stem cells in the first reduced image is less than a preset first density, calculating at least one density of stem cells in the cell image according to a plurality of preset second ratios and a plurality of pre-trained second density calculation models, wherein the plurality of preset second ratios are in one-to-one correspondence with the plurality of pre-trained second density calculation models. ### Claim 16 (depends on 15) The non-transitory storage medium according to claim 15, the method further comprising: when the at least one density of stem cells in the cell image is less than a preset second density, calculating a density of stem cells in the cell image by using a pre-trained third density calculation model. ### Claim 17 (depends on 15) The non-transitory storage medium of claim 15, wherein calculating at least one density of stem cells in the cell image according to a plurality of preset second ratios and a plurality of pre-trained second density calculation models comprises: obtaining a largest second ratio among the plurality of preset second ratios; reducing the image size of the cell image according to the largest second ratio and obtaining a second reduced image; calculating a density of stem cells in the second reduced image by using a pre-trained second density calculation model corresponding to the largest second ratio; when the density of stem cells in the second reduced image is less than a preset second density corresponding to the largest second ratio, obtaining a second largest second ratio among the plurality of preset second ratios; reducing the image size of the cell image according to the second largest second ratio and obtaining a third reduced image; and calculating a density of stem cells in the third reduced image by using a pre-trained second density calculation model corresponding to the second largest second ratio. ### Claim 18 (depends on 17) The non-transitory storage medium of claim 17, the method further comprising: if the density of stem cells in the second reduced image is greater than or equal to the preset second density corresponding to the largest second ratio, determining that the density of stem cells in the cell image is greater than or equal to the preset second density corresponding to the largest second ratio but less than the preset first density. ### Claim 19 (depends on 18) The non-transitory storage medium of claim 18, before acquiring the cell image, the method further comprising: acquiring a plurality of first training images; selecting, from the plurality of first training images, images with a density of stem cells greater than or equal to the preset first density as first positive sample images, and images with a density of stem cells less than the preset first density as first negative sample images; reducing image sizes of the first positive sample images and image sizes of the first negative sample images according to the preset first ratio; and training and obtaining the first density calculation model with the reduced first positive sample images and the reduced first negative sample images. ### Claim 20 (depends on 19) The non-transitory storage medium of claim 19, before acquiring the cell image, the method further comprising: acquiring a plurality of second training images; for a target second ratio of the plurality of preset second ratios, selecting, from the plurality of second training images, images with a density of stem cells greater than or equal to a preset second density corresponding to the target second ratio as second positive sample images, and images with a density of stem cells less than the preset second density corresponding to the target second ratio as second negative sample images; reducing image sizes of the second positive sample images and image sizes of the second negative sample images according to the target second ratio; and training and obtaining a second density calculation model corresponding to the target second ratio with the reduced second positive sample images and the reduced second negative sample images.