Digital X-ray Image Processing for Low Information Region Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Clinical X-ray images often contain regions with low information content, such as direct radiation areas or shadows of radiation-opaque objects, which complicate the visualization and window level and-width adjustment procedures, making it difficult to accurately render anatomical images.

Innovation Solution

An image processing system that filters input images to create a structure image, identifies regions of interest using histogram analysis, and computes weights for image values outside these regions to distinguish between regions of interest and low information content, allowing for improved visualization by excluding low information regions from brightness and contrast calculations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If histogram-based window level and-width adjustment is performed on the entire image, then brightness and contrast can be adjusted, but regions with low information content (such as direct radiation areas or shadows of radiation-opaque objects) interfere with the adjustment and make it difficult to achieve accurate visualization of anatomical structures

Engineering Contradiction:

Improveaccuracy of brightness and contrast adjustmentVSAvoidcomplexity of window level and-width adjustment procedure

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The image is segmented into multiple regions based on intensity values: a first region corresponding to anatomical structures of interest, a second region corresponding to direct radiation areas, and a third region corresponding to other structures. This segmentation allows different processing to be applied to each region, enabling accurate window level and-width adjustment for anatomical structures while excluding low information content regions from the adjustment calculation.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Regions with low information content (direct radiation areas and shadows of radiation-opaque objects) are identified and extracted from the image based on intensity thresholds. These extracted regions are then excluded from the histogram-based window level and-width adjustment calculation, preventing them from interfering with the brightness and contrast adjustment of anatomical structures.

Inventive Principle:

Principle #2Taking out (Extraction)

2Reliability

If regions with low information content are included in the image analysis, then the complete image is processed, but the visualization quality deteriorates due to interference from non-diagnostic regions

Engineering Contradiction:

Improvequality of anatomical image visualizationVSAvoidamount of image data processed

Core Design Contradiction:

ReliabilityVSQuantity of substance

Solution Approach 1:

The image is divided into multiple intensity-based regions, separating anatomical structures of interest from low information content areas. This segmentation enables selective processing where only relevant regions contribute to the visualization quality, improving reliability while reducing the effective quantity of data that needs to be processed for diagnostic purposes.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Low information content regions are identified through intensity thresholding and extracted from the image data. By removing these regions from the analysis, the system processes only the quantity of data that contributes to diagnostic quality, improving visualization reliability without unnecessarily processing non-diagnostic areas.

Inventive Principle:

Principle #2Taking out (Extraction)

3Reliability

If manual window level and-width adjustment is performed to compensate for low information content regions, then visualization can be improved, but the procedure becomes more cumbersome and time-consuming

Engineering Contradiction:

Improvequality of anatomical structure renderingVSAvoidtime required for brightness and contrast adjustment

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system automatically performs window level and-width adjustment by identifying regions of interest and calculating optimal brightness and contrast parameters based on the histogram of those regions. This self-service approach eliminates the need for manual adjustment by radiologists, reducing time loss while maintaining high rendering quality through automated intelligent processing.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

By automatically segmenting the image into anatomical structures of interest and low information content regions, the system enables automated window level and-width adjustment specific to the anatomical regions. This eliminates the need for manual trial-and-error adjustment, significantly reducing the time required while improving rendering quality through region-specific optimization.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS11354806B2Detection of regions with low information content in digital X-ray images

Publication Date: 2022.06.07 KONINKLIJKE PHILIPS NV

AI summary

An image processing system and related method. The system comprises an input interface (IN) configured for receiving an input image. A filter (FIL) of the system filters said input image to obtain a structure image from said input image, said structure image including a range of image values. A range identifier (RID) of the system identifies, based on an image histogram for the structure image, an image value sub-range within said range. The sub-range being associated with a region of interest. The system output through an output interface (OUT) a specification for said image value sub-range. In addition or instead, a mask image for the region of interest or for region or low information is output.