Medical Image Enhancement via Selective Non-Linear Filtering

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Medical image datasets acquired during surgical interventions, particularly minimally invasive procedures, are challenging to interpret due to the presence of anatomical structures and medical devices like stents being close to metal coils or strongly attenuating structures, leading to difficulties in visualizing high-contrast objects like medical devices.

Innovation Solution

A method involving non-linear filtering is applied selectively to image regions within a predetermined image-value interval, generating an enhancement dataset that, when added to the original image dataset, enhances the visibility of medical devices while avoiding noise amplification and artifacts, by using a non-linear low-pass filter to exclude untargeted image regions and retain only the object of interest.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Difficulty of detecting and measuring

If non-linear filtering is applied to the entire image dataset, then the visibility of medical devices is improved, but noise and artifacts are amplified throughout the image

Engineering Contradiction:

Improvevisibility of medical devicesVSAvoidnoise amplification

Core Design Contradiction:

Difficulty of detecting and measuringVSObject-affected harmful factors

Solution Approach 1:

The patent applies non-linear filtering selectively only to image regions where image values fall within a predetermined interval corresponding to the medical device of interest. This local quality approach ensures that enhancement is applied only where needed, improving device visibility while avoiding noise amplification in other regions such as anatomical structures or areas with strong attenuators.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The image dataset is segmented into different regions based on image-value intervals. The method identifies and isolates the specific interval corresponding to the medical device, then applies filtering only to that segment. This segmentation allows targeted enhancement without affecting the entire image, thereby reducing overall noise and artifact amplification.

Inventive Principle:

Principle #1Segmentation

2Difficulty of detecting and measuring

If conventional filtering methods are used, then image enhancement is achieved, but surrounding anatomical structures and medical aids are also amplified

Engineering Contradiction:

Improveenhancement of medical devicesVSAvoiddistortion of surrounding structures

Core Design Contradiction:

Difficulty of detecting and measuringVSLoss of information

Solution Approach 1:

The patent employs local quality by restricting the non-linear filtering operation to a specific image-value interval that corresponds to the medical device. This ensures that only the desired object is enhanced while surrounding anatomical structures and medical aids, which have different image values, remain unaffected and are not erroneously amplified.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The predetermined image-value interval acts as an intermediary selector that mediates between the original image data and the filtered enhancement. By using this intermediate criterion, the method selectively identifies and processes only the regions corresponding to the medical device, preventing distortion of surrounding structures while achieving the desired enhancement.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Measurement precision

If image processing is applied to improve medical device visibility, then diagnostic accuracy is improved, but processing time and computational complexity increase

Engineering Contradiction:

Improvediagnostic accuracyVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

Instead of applying filtering to the entire image dataset (excessive action), the patent applies the computationally intensive non-linear filtering only to the partial region where image values fall within the predetermined interval. This partial action significantly reduces processing time and computational complexity while maintaining diagnostic accuracy for the medical device of interest.

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The method performs preliminary identification of the image-value interval corresponding to the medical device before applying the non-linear filtering. This preliminary action allows the system to target only the relevant regions for processing, reducing overall computation time while ensuring diagnostic accuracy is not compromised.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11599977B2Image-processing of image datasets of patients

Publication Date: 2023.03.07 SIEMENS HEALTHINEERS AG

AI summary

A method is provided for image-processing an image dataset acquired from a patient by a medical imaging apparatus, (e.g., an X-ray apparatus), wherein the image dataset includes image values associated with image points, and depicts an acquisition region of the patient containing at least one object, (e.g., a medical device), to be enhanced, which is represented by image values within an image-value interval. The method includes determining a non-linearly high-pass filtered enhancement dataset, which is confined to an image portion containing image values lying in the image-value interval. The method also includes determining a result dataset by adding to the image dataset the enhancement dataset weighted by a weighting value. The method further includes outputting the result dataset.