Digital Slide Quality Assessment via Spatial Frequency Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In digital pathology, assessing image quality is challenging due to the subjective nature of evaluating complex biological specimens, with existing methods failing to reliably distinguish between good and bad quality images, especially in the presence of scan-related artifacts.

Innovation Solution

A system that analyzes digital slide images based on complexity and spatial frequencies, using a two-part analysis process (micro-analysis and macro-analysis) to generate a quantifiable assessment, providing both visual and quantitative feedback through a color-coded heat map, thereby identifying areas of poor quality and scan-related artifacts.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual review of digital slide images is performed, then quality assessment can be conducted, but it is time consuming and requires expert judgment which is highly subjective

Engineering Contradiction:
Improvequality assessment accuracyVSAvoidreview time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces the manual mechanical review process with an automated computer-based image analysis system that uses algorithms to objectively assess digital slide quality, eliminating the need for time-consuming expert visual inspection while maintaining or improving assessment accuracy

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service quality assessment by allowing the digital slide imaging system to automatically evaluate its own output quality through integrated analysis algorithms, without requiring external expert review

Inventive Principle:
Principle #25Self-service

2Productivity

If automated image quality metrics are used, then review time is reduced, but existing methods fail to reliably distinguish between good and bad quality images

Engineering Contradiction:
Improvereview efficiencyVSAvoidquality distinction reliability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent combines multiple image quality metrics into a composite quality score, integrating various measurement dimensions (sharpness, contrast, noise, artifacts) to create a comprehensive and reliable quality assessment that overcomes the limitations of single metrics

Inventive Principle:
Principle #40Composite materials

Solution Approach 2:

The system dynamically adjusts and optimizes multiple image quality parameters simultaneously, using weighted combinations of different metrics that can be tailored to specific application requirements, thereby improving both reliability and adaptability

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If expert judgment is used for quality assessment, then subjective variations are minimized, but the process remains highly subjective and inconsistent

Engineering Contradiction:
Improvequality judgment consistencyVSAvoidassessment system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces the human expert judgment system with an automated computational system that applies consistent algorithms and objective criteria, eliminating subjective variations while maintaining assessment precision through mathematically defined quality metrics

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP2577602B1System and method to determine slide quality of a digitized microscope slide
Publication Date: 2019.11.27 LEICA BIOSYSTEMS IMAGING INC
  • EP2577602B1 patent drawingFigure 1~2B
  • EP2577602B1 patent drawingFigure 3~4B
  • EP2577602B1 patent drawingFigure 5A

AI summary

A system that determines the quality of a digital microscope slide by analyzing digital slide images based on complexity and spatial frequencies. An example embodiment detailed in the application may provide visual feedback on the whole slide quality by overlaying the image with a color coded "heat map" of local area quality. A user provided with the overlap image may obtain both an absolute quality measurement for the whole image and quickly identity the quality variability within the slide.