IVF Dish Drop Tracking Using Computer Vision Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In the in-vitro fertilization (IVF) process, maintaining accurate tracking of biological materials during embryo handling is challenging due to the manual labeling and witnessing requirements, which are prone to errors and increase costs.
Innovation Solution
An automated tracking system using computer vision and machine learning to identify and assign unique identifiers to biological materials and their locations, reducing the need for manual witnessing and improving the accuracy of transfers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual labeling and witnessing procedures are used to track biological materials, then accurate tracking can be maintained, but the process is prone to human errors and increases operational costs
Solution Approach 1:
The patent replaces manual mechanical tracking processes with an automated computer vision system. Cameras capture images of dishes and drops, machine learning models identify and track biological materials, and the system automatically generates reports. This substitution eliminates human error in tracking while reducing operational complexity through automation.
Solution Approach 2:
The system creates digital copies of the tracking process through automated image capture and analysis. Instead of manual visual inspection and recording, the system captures images of dishes and drops, processes them through machine learning models to create digital representations of the biological materials and their locations, and maintains automated records of all transfers and movements.
2Reliability
If a second embryologist witnesses each transfer step to prevent mistakes, then tracking accuracy improves, but the time required for each operation increases
Solution Approach 1:
The system enables self-service tracking where the computer vision system automatically monitors and records all transfers without requiring a second embryologist. The machine learning models independently identify biological materials, track their movements between dishes and drops, and generate accuracy reports, freeing up embryologists to focus on hands-on work without time loss to witnessing procedures.
3Productivity
If multiple drops are placed on a single dish, then the workflow efficiency improves, but the difficulty of detecting and measuring specific drops increases
Solution Approach 1:
The system segments the complex task of tracking multiple drops on a dish by first identifying the dish as a whole, then dividing it into individual drop regions. The machine learning model processes the image to detect each drop's location, size, and characteristics independently, allowing precise identification and tracking of each drop even when multiple drops are present on the same dish.
Solution Approach 2:
The system applies local quality analysis by examining specific regions of the dish image corresponding to individual drops. The machine learning model analyzes local characteristics of each drop (such as shape, size, and position) to identify and track them separately, enabling precise measurement and monitoring of each drop's contents and movements within the context of the entire dish.
Data Source
AI summary
The present disclosure relates to a method performed by one or more computers for tracking a biological material of a subject during an in-vitro fertilization process. The method includes receiving, from a camera, an image of a dish having a visual characteristic and a drop disposed on the dish, the dish holding the biological material at a drop location. The method then includes processing the image of the dish, using a drop identification model, to identify the drop according to the visual characteristic. Further, the method includes assigning an identifier to the drop associated with the drop location, and recording the identifier of the drop associated with the drop location.