Multi-Focal Plane Imaging With Selective ROI Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for multi-focal plane image acquisition in digital pathology require large storage capacities and do not allow pathologists to manually control focal planes for regions of interest, leading to inefficient data management and storage needs.
Innovation Solution
Adaptive scanning and multi-scale wavelet transforms are used to capture images at varying resolutions based on contrast metrics, reducing storage requirements by selectively capturing high-resolution images of regions of interest while maintaining sufficient resolution for analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complete multi-focal plane images are captured for all regions, then sufficient analysis resolution is maintained, but storage requirements increase significantly
Solution Approach 1:
The patent applies local quality by capturing high-resolution images only for regions of interest (ROI) identified by pathologists, while capturing lower-resolution images for non-critical regions. This selective resolution approach maintains sufficient analysis quality for important areas while reducing overall storage requirements for the complete slide image.
Solution Approach 2:
The patent segments the slide image into multiple focal planes and further divides regions into ROI and non-ROI areas. By segmenting the image data structure into different resolution levels and focal planes, the system allows selective storage and retrieval of high-resolution data only where needed, optimizing the balance between analysis precision and storage efficiency.
2Adaptability or versatility
If all focal planes are captured at high resolution, then comprehensive analysis capability is achieved, but data management complexity increases
Solution Approach 1:
The patent implements a dynamic data management system where the resolution and focal plane selection adapt based on user interaction. Pathologists can dynamically request high-resolution views of specific regions at specific focal planes during analysis, rather than having all high-resolution data statically stored. This dynamic approach maintains comprehensive analysis capability while simplifying data management through on-demand retrieval.
Solution Approach 2:
The system performs preliminary low-resolution scanning of entire slides to identify potential regions of interest before committing to high-resolution capture. This preliminary action allows the system to prepare data structures in advance that can efficiently accommodate high-resolution data only where needed, reducing overall data management complexity while maintaining analysis versatility.
3Measurement precision
If high-resolution images are captured for all regions, then detailed feature detection is improved, but storage requirements increase
Solution Approach 1:
The patent applies local quality by capturing high-resolution images only for regions of interest (ROI) identified by pathologists, while capturing lower-resolution images for non-critical regions. This selective resolution approach maintains sufficient analysis quality for important areas while reducing overall storage requirements for the complete slide image.
Data Source
AI summary
Example methods and apparatus to perform multi-focal plane image acquisition and compression are disclosed. A disclosed example method includes capturing a first image of a portion of an object at a first focal plane and at a first resolution, computing a contrast metric for the captured first image, comparing the contrast metric to a threshold to determine whether to capture a second image of the portion of the object at the first focal plane and at a second resolution, wherein the second resolution is different from the first resolution, capturing the second image of the portion of the object at the first focal plane and at the second resolution, and storing a first representation of the second image in a file, the file containing a second representation of the portion of the object at a second focal plane, wherein the second representation is at the first resolution.