Automated Seeding of Cytological Images for Reviewer Vigilance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The difficulty in maintaining and measuring cytotechnologist diligence during cytological slide examinations, particularly due to low prevalence of abnormal cases in Pap smear slides, leads to challenges in accurately assessing sensitivity and vigilance, as conventional seeding methods are tedious and inefficient.
Innovation Solution
A computer-assisted method that analyzes images of cytological samples, identifies objects of interest, and intersperses images of previously classified objects into the workflow to maintain a threshold rate of abnormal cases, enhancing reviewer alertness and accuracy through automated seeding of digital images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional seeding methods are used to maintain reviewer alertness, then reviewer vigilance can be improved, but the process becomes tedious and inefficient requiring excessive slide handling
Solution Approach 1:
The patent uses digital copies of previously classified abnormal cases stored in a database. Instead of physically handling and disguising actual abnormal slides, the system retrieves and displays digital images of previously identified abnormal cases from the database, eliminating the need for physical slide manipulation while maintaining reviewer alertness
Solution Approach 2:
The patent replaces the mechanical process of physically selecting, disguising, and handling abnormal slides with an automated computer-based system that electronically retrieves and displays images from a database, significantly reducing manual intervention and improving efficiency
2Quantity of substance
If the prevalence of abnormal cases is low in Pap smear slides, then more resources can be allocated to other tasks, but it becomes difficult to maintain and measure cytotechnologist diligence and sensitivity
Solution Approach 1:
The patent implements a feedback mechanism where the system monitors the reviewer's performance in real-time and dynamically adjusts the seeding of abnormal cases. The system tracks metrics such as detection rate and response time, providing feedback to maintain optimal reviewer alertness while ensuring accurate measurement of sensitivity and diligence
Solution Approach 2:
The patent performs preliminary classification and storage of abnormal cases in a database before they are needed for seeding. This allows the system to have pre-prepared images ready for immediate retrieval and display, enabling efficient monitoring and measurement of reviewer performance without requiring real-time physical manipulation of slides
Data Source
AI summary
A computer-assisted method of classifying cytological samples, includes using a processor to analyze images of cytological samples and identify cytological objects of interest within the sample images, wherein the processor (i) displays images of identified cytological objects of interest from the sample images to a reviewer, (ii) accesses a database of images of previously classified cytological objects, and (iii) displays to the reviewer, interspersed with the displayed images of the identified objects of interest from the sample images, one or more images obtained from the database of images of previously-classified objects.