Medical Imaging Result Generation with Adjustable Processing Strength

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing medical imaging algorithms often introduce algorithmic artifacts and make it difficult for observers to verify the accuracy of processed images, especially when using artificial intelligence, leading to potential misinterpretation due to insufficient image information and undetectable deviations from the original raw image.

Innovation Solution

A computer-implemented method that allows users to adjust an overall processing algorithm through a control parameter, enabling continuous or staged variation of parameter settings for part algorithms, allowing users to assess how much and how the result image data deviates from the original data, and adapt the processing strength accordingly.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If image processing algorithms are applied to modify the raw image, then the image is prepared better for the intended usage purpose, but the processed image removes data from the raw image and introduces algorithmic artifacts

Engineering Contradiction:

Improveimage quality for intended usageVSAvoiddata removal from raw image

Core Design Contradiction:

ReliabilityVSLoss of information

Solution Approach 1:

The patent implements a feedback mechanism by displaying both the processed result image and the original raw image side by side, allowing the observer to compare and verify the processing effects. This feedback loop enables the observer to assess whether the processing has introduced unwanted artifacts or removed important information, thus resolving the contradiction between improving image quality and preserving original data integrity.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent segments the image processing workflow into distinct visible components: the original raw image and the processed result image. By presenting these as separate but comparable entities, the system allows observers to evaluate the processing effects independently while maintaining the ability to assess the relationship between original and processed data, thereby balancing reliability improvement with information preservation.

Inventive Principle:

Principle #1Segmentation

2Adaptability or versatility

If artificial intelligence image processing algorithms are used to interpret image contents, then the image processing capability is enhanced, but the results are not verifiable and familiar image properties may become lost

Engineering Contradiction:

Improveimage interpretation capabilityVSAvoidverifiability of processing results

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent creates a visual copy of the original raw image alongside the AI-processed result image. This copying approach allows observers to verify the processing results by comparing them against the original unprocessed data, maintaining measurement precision and verifiability even when using advanced AI interpretation algorithms that enhance adaptability and versatility.

Inventive Principle:

Principle #26Copying

3Manufacturing precision

If interpreting algorithms are applied to free noisy images from noise, then the image appears cleaner, but the impression of high-dose image may arise that does not reflect the real information content

Engineering Contradiction:

Improvenoise reduction qualityVSAvoidreal information content representation

Core Design Contradiction:

Manufacturing precisionVSLoss of information

Solution Approach 1:

The side-by-side display of processed and original images provides immediate feedback to observers about the extent of noise reduction and potential information loss. This visual feedback mechanism prevents misleading impressions by allowing direct comparison, ensuring that cleaned images do not falsely represent high-dose quality when the real information content may be compromised.

Inventive Principle:

Principle #23Feedback

4Reliability

If multiple image processing stages are applied, then the image is better prepared for usage, but the complexity of the processing algorithm increases

Engineering Contradiction:

Improveimage preparation qualityVSAvoidprocessing algorithm complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The visual feedback system allows observers to assess the cumulative effects of multiple processing stages by comparing the final result with the original image. This transparency reduces the perceived complexity by making the processing effects visible and evaluable, even though the underlying algorithm remains complex. The feedback mechanism builds trust in the multi-stage processing by demonstrating its effects.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20250336201A1Generation of result image data from origin data based on a medical imaging

Publication Date: 2025.10.30 SIEMENS HEALTHINEERS AG

AI summary

A method for generating result image data from origin data includes a part algorithm generated as a function of input data and a parameter setting generated as output data. The parameter setting of at least one part algorithm is predetermined as a function of a control parameter of an overall processing algorithm. First and second limit values are assigned first and second values of the parameter setting, wherein that of the first value for the at least one part algorithm leads to the output data of the at least one part algorithm deviating more greatly from the input data of that of the part algorithms or from reference data, which would result on application of a reference algorithm assigned to the at least one part algorithm to this input data, than with using the second value of the parameter setting of the respective part algorithm.