# US 2025/0095150 A1 · Image Processing for Real-Time Ranking and Selection of Spermatozoa > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2025/0095150 A1 - Document ID: US20250095150A1 - Status: Published application - Kind: A1 - Application number: 18/965,245 - Filed: 2024-12-02 - Published: 2025-03-20 - Earliest claimed: 2020-11-17 - Assignees: Conceivable Life Sciences Inc. - Applicants: Conceivable Life Sciences Inc. - Inventors: Alejandro Chávez Badiola - Web page: https://patentgrep.com/patent/US20250095150A1 ## CPC - G06T 7/0012: {Biomedical image inspection} - G06T 7/254: involving subtraction of images - G06T 2207/10016: Video; Image sequence - G06T 2207/10056: Microscopic image - G06T 2207/20081: Training; Learning - G06T 2207/20084: Artificial neural networks [ANN] - G06T 2207/30024: Cell structures in vitro; Tissue sections in vitro - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL ## Abstract A method includes capturing multiple images over time of objects in a sperm sample and using image processing to identify a set of candidate spermatozoa from the objects. The image processing includes identifying the set of candidate spermatozoa based on movement characteristics of the set of candidate spermatozoa observed over time. The method includes, for each respective spermatozoon of the set of candidate spermatozoa, determining a quality metric for the respective spermatozoon based on trajectory characteristics of the respective spermatozoon indicative of motility. The method includes ranking the set of candidate spermatozoa based on the quality metric. The method includes selecting a highest-ranked spermatozoon from the ranking. The method includes displaying at least a portion of the sperm sample on a screen with a visual indicator identifying the selected spermatozoon. The method includes designating the selected spermatozoon for extraction from the sperm sample and attempted fertilization. ## Claims ### Claim 1 A method comprising: capturing a plurality of images over time of a set of objects in a sperm sample; using image processing to identify a set of candidate spermatozoa from the set of objects, wherein the image processing includes identifying the set of candidate spermatozoa from the set of objects based on movement characteristics of the set of candidate spermatozoa observed over time; for each respective spermatozoon of the set of candidate spermatozoa, determining a quality metric for the respective spermatozoon based on trajectory characteristics of the respective spermatozoon indicative of motility; ranking the set of candidate spermatozoa based on the quality metric; selecting a highest-ranked spermatozoon from the ranking; displaying at least a portion of the sperm sample on a screen with a visual indicator identifying the selected spermatozoon; and designating the selected spermatozoon for extraction from the sperm sample and attempted fertilization. ### Claim 2 (depends on 1) The method of claim 1 further comprising displaying on the screen a second visual indicator identifying a next-highest-ranked spermatozoon from the ranking. ### Claim 3 (depends on 1) The method of claim 1 further comprising extracting the selected spermatozoon from the sperm sample for attempted fertilization. ### Claim 4 (depends on 3) The method of claim 3 wherein attempted fertilization is performed using intracytoplasmic sperm injection (ICSI). ### Claim 5 (depends on 1) The method of claim 1 wherein the trajectory characteristics include a speed of rotation about a longitudinal axis. ### Claim 6 (depends on 5) The method of claim 5 wherein the trajectory characteristics include at least one of linearity of movement or speed of movement. ### Claim 7 (depends on 1) The method of claim 1 wherein the quality metric for a respective spermatozoon is based on a movement pattern of the respective spermatozoon. ### Claim 8 (depends on 7) The method of claim 7 wherein the movement pattern includes movement of a head portion of the respective spermatozoon relative to a remainder of the respective spermatozoon. ### Claim 9 (depends on 7) The method of claim 7 wherein the movement pattern includes movement of a tail portion of the respective spermatozoon relative to a remainder of the respective spermatozoon. ### Claim 10 (depends on 1) The method of claim 1 wherein the visual indicator includes a graphic element, overlaid on at least one of the plurality of images, that indicates a location of the selected spermatozoon. ### Claim 11 (depends on 1) The method of claim 1 wherein the movement characteristics of the set of candidate spermatozoa are determined within 500 milliseconds of the plurality of images being captured. ### Claim 12 (depends on 1) The method of claim 1 wherein: the image processing includes tracking movement of the set of candidate spermatozoa across the plurality of images using artificial intelligence, and the artificial intelligence includes a convolutional neural network (CNN). ### Claim 13 (depends on 1) The method of claim 1 wherein the image processing includes determining shape characteristics of the set of objects. ### Claim 14 (depends on 13) The method of claim 13 wherein the image processing includes applying digital filters to compare the shape characteristics to reference patterns. ### Claim 15 (depends on 13) The method of claim 13 wherein the shape characteristics include at least two of dimension, area, eccentricity, height, width, or convexity. ### Claim 16 (depends on 1) The method of claim 1 wherein: the quality metric for a respective spermatozoon is based on a set of shape characteristics of the respective spermatozoon, and the set of shape characteristics includes at least one of head shape, head size, tail size, and presence of anomalous shape features. ### Claim 17 (depends on 1) The method of claim 1 wherein the quality metric for a respective spermatozoon is based on texture of the respective spermatozoon. ### Claim 18 (depends on 1) The method of claim 1 wherein: the quality metric is determined using a machine learning model, and features input into the machine learning model include the trajectory characteristics, the movement characteristics, and shape characteristics. ### Claim 19 A system comprising: a camera configured to capture a plurality of images over time of a set of objects in a sperm sample; a display; and a processing unit configured to: use image processing to identify a set of candidate spermatozoa from the set of objects, wherein the image processing includes identifying the set of candidate spermatozoa from the set of objects based on movement characteristics of the set of candidate spermatozoa observed over time, determine a quality metric for each respective spermatozoon of the set of candidate spermatozoa based on trajectory characteristics of the respective spermatozoon indicative of motility, rank the set of candidate spermatozoa based on the quality metric, select a highest-ranked spermatozoon from the ranking, present, on the display, at least a portion of the sperm sample on a screen with a visual indicator identifying the selected spermatozoon, and designate the selected spermatozoon for extraction from the sperm sample and attempted fertilization. ### Claim 20 (depends on 19) The system of claim 19 wherein the camera captures the plurality of images at a magnification of at least 20×.