Surgical Instrument Tracking via Machine Vision Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Surgical device reprocessing in hospitals faces challenges due to inadequate tracking of complex processes, leading to issues like loss of productivity, mismanagement of equipment, and potential healthcare-associated infections, with machine vision techniques struggling with large datasets and confidence issues in identifying surgical instruments.
Innovation Solution
A system incorporating an image capture device and a computing device that uses machine vision techniques to identify and track surgical instruments, including determining sterilization status and performing operations based on flagged events, with features like optically active articles and background surfaces to enhance identification accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine vision techniques are used to identify individual surgical instruments, then identification capability is improved, but computing resources required increase significantly when handling large data sets
Solution Approach 1:
The system segments the identification process by using multiple image capture devices positioned at different locations (sterile storage area, sterilization area, distribution area) to capture images of surgical instruments at various stages. This distributes the processing load across multiple localized systems rather than requiring one centralized system to process all images, reducing the computing resource burden on any single system while maintaining identification capability.
2Extent of automation
If machine vision techniques are used to identify surgical instruments, then automated tracking is improved, but confidence values and classification accuracy deteriorate
Solution Approach 1:
The system implements feedback mechanisms where captured images are processed and compared against known instrument characteristics. The processor provides feedback by determining whether captured images match expected instrument patterns, and can trigger alerts or notifications when confidence levels are insufficient. This feedback loop allows the system to self-correct and verify classifications, improving reliability while maintaining automated tracking.
Solution Approach 2:
The system performs preliminary actions by capturing images at multiple predetermined locations (sterile storage area, sterilization area, distribution area) before instruments are used. This creates a chain of custody record with pre-established confidence levels at each stage, allowing the system to verify instrument identity multiple times throughout the process rather than relying on a single classification event.
3Loss of information
If multiple image capture devices are deployed to track instruments through multiple areas, then tracking completeness is improved, but device complexity increases
Solution Approach 1:
The system uses universal image capture devices that can function in multiple areas (sterile storage area, sterilization area, distribution area) rather than requiring specialized devices for each location. Each device performs the same basic function of capturing instrument images, but the system integrates data from all locations to achieve complete tracking. This multi-functionality approach reduces overall system complexity while maintaining tracking completeness.
4Productivity
If automated operations are performed based on instrument identification, then productivity is improved, but risk of errors increases
Solution Approach 1:
The system performs preliminary verification by capturing and analyzing images at multiple stages (sterile storage area, sterilization area, distribution area) before automated operations are executed. This preliminary action creates multiple verification points where instrument identity and sterilization status can be confirmed, reducing the risk of errors in automated operations while maintaining productivity improvements.
Data Source
AI summary
Aspects of the present disclosure relate to a system including an image capture device and a computing device configured to receive a test image from the image capture device corresponding to a first surgical instrument, determine an identity type of the first surgical instrument using the test image in a machine vision technique, determine whether the first surgical instrument is flagged, and perform at least one operation in response to whether the first surgical instrument is flagged.