Retained Surgical Item Detection via Digital X-Ray Catalog
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting retained surgical items (RSI) during surgery are inefficient, with poor detection rates due to the use of portable radiography and lack of a reference catalog of x-ray images, leading to difficulties in identifying small or broken items amidst a cluttered surgical field.
Innovation Solution
A system utilizing a portable digital x-ray machine and a database with images and measurements of surgical items, allowing for comparison of patient images to identify retained items through automated shape recognition and machine vision techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If portable radiography is used to detect retained surgical items, then the detection method is simple and quick, but the detection accuracy is poor (33-50%)
Solution Approach 1:
The system transitions from single-plane portable radiography to multi-planar digital imaging (AP, lateral, and oblique views), adding dimensional information to improve detection accuracy while maintaining efficiency through automated analysis
Solution Approach 2:
A computer-aided detection system with machine learning algorithms acts as an intermediary between the radiologist and the images, automatically identifying and flagging potential retained items to improve detection accuracy without significantly increasing workflow time
2Measurement precision
If a reference catalog of x-ray images with surgical item sizes is created, then the ability to identify retained items improves, but the system complexity increases
Solution Approach 1:
The system creates a digital reference catalog containing standardized images and measurements of common surgical items (clips, sponges, hemostats, etc.), which can be automatically compared against patient images to identify retained items, improving accuracy through pattern recognition
Solution Approach 2:
The reference catalog is pre-populated with images and measurements of surgical items before the surgical procedure begins, allowing for rapid comparison and identification during the procedure without adding complexity to the actual detection process
3Measurement precision
If multiple imaging views and automated analysis are implemented, then the detection accuracy improves, but the time required for imaging and analysis increases
Solution Approach 1:
The automated detection system processes multiple imaging views simultaneously and continuously monitors for retained items throughout the surgical procedure, eliminating the need for separate analysis sessions and reducing total time investment
Solution Approach 2:
The system provides real-time feedback by automatically analyzing images as they are acquired and immediately flagging potential retained items, allowing for rapid decision-making without requiring time-consuming manual review of all images
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances the accuracy and efficiency of detecting retained surgical items, reducing the risk of complications and operating time, while aligning with existing protocols and reducing medico-legal exposure.
Implementation Method 1
obtaining intra-operative surgical image data of the patient with a portable digital x-ray machine
Data Source
AI summary
The present disclosure is directed, in part, to methods and systems of intra-operatively identifying a retained surgical item by imaging a surgical item with an intra-operative imaging device to obtain image data, imaging a patient suspected of having a retained surgical item with an intra-operative imaging device to obtain image data, interacting the image data to a module configured to identify the surgical item from the intra-operative surgical item image data, and identifying the retained surgical item from the surgical item image data.