Decoupled Imaging and Analysis Architecture for Slide Throughput
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing slide processing systems are bottlenecked by the shared computer resource constraint, where the computer used for imaging also performs analysis, leading to slowed processing times and reduced throughput, especially with dense slides, and advanced analysis algorithms are impractical due to limited resources.
Innovation Solution
Decoupling image acquisition and analysis by using separate imaging and analysis computers connected through a network, allowing analysis computers to process image data independently and efficiently, eliminating the need for a single shared computer to handle both tasks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single shared computer is used for both image acquisition and analysis, then device complexity is reduced, but processing speed and throughput deteriorate due to resource constraints
Solution Approach 1:
The system divides the previously unified computer function into two separate computing systems: an imaging computer for acquiring images and an analysis computer for processing images. This segmentation allows each system to specialize in its function, eliminating resource contention and improving overall throughput while maintaining manageable system complexity through clear functional separation.
Solution Approach 2:
The image analysis function is extracted from the imaging computer and assigned to a dedicated analysis computer. This extraction removes the bottleneck caused by shared resources, allowing the imaging computer to focus solely on acquisition while the analysis computer handles processing, thereby increasing slide throughput without significantly increasing overall system complexity.
2Measurement precision
If advanced analysis algorithms are implemented on the shared computer, then measurement precision and analysis accuracy improve, but processing time increases and throughput decreases
Solution Approach 1:
Complex analysis algorithms are extracted from the resource-constrained imaging computer and executed on a dedicated analysis computer with sufficient computational resources. This allows sophisticated algorithms to run at full speed without compromising imaging performance, achieving both high analysis accuracy and maintained throughput.
Solution Approach 2:
A network connection acts as an intermediary between the imaging computer and analysis computer, allowing advanced algorithms to be executed remotely on the analysis computer while the imaging computer continues its acquisition function. This mediator enables complex processing without creating bottlenecks in the image acquisition pipeline.
3Loss of time
If the computer keeps up with real-time analysis during image acquisition, then processing time per slide is minimized, but system adaptability and flexibility deteriorate
Solution Approach 1:
Images are acquired and stored in advance on the imaging computer or directly on the network, allowing the analysis computer to process them at its own pace without real-time pressure. This preliminary action separates the acquisition phase from the analysis phase, enabling flexible scheduling and multiple processing modes while maintaining efficient throughput.
Solution Approach 2:
The system transitions from rigid real-time processing to a dynamic architecture where the analysis computer can adjust its processing speed and methods based on workload, slide complexity, and resource availability. This dynamic approach allows the system to adapt to varying requirements while maintaining overall efficiency through the decoupled architecture.
Data Source
AI summary
Method and system for imaging and analyzing a biological specimen on a specimen carrier, such as a slide. One or more imagers acquire images of a biological specimen on a slide and generate electronic image data. One or more analysis computers, such as a cluster of analysis computers, are connected to the imagers through a network and process the electronic image data. The number of analysis computers can be different than the number of imagers, and analysis computers can be located remotely from the imagers. The results of processing by the analysis computers are stored to a database, which is accessible by one or more review stations.