Medical Image Reconstruction with Adaptive Parameter Optimization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current medical imaging technologies face challenges in efficiently reconstructing patient-specific medical image data sets, particularly in screening examinations where relevant structures are unknown, leading to suboptimal image quality and increased storage and processing demands.

Innovation Solution

A method and system that utilize a preliminary image data set reconstructed with default parameters, analyzed by an algorithm to determine analysis information, and then re-reconstructed using a second parameter set tailored to the individual patient, optimizing image content and reducing the need for manual re-processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If standard reconstruction parameter sets are used for screening examinations, then processing time is reduced and throughput is increased, but image quality and diagnostic accuracy deteriorate because relevant structures are not optimized

Engineering Contradiction:
ImprovethroughputVSAvoidimage quality
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The system performs a preliminary reconstruction with default parameters to generate a preliminary image data set, then uses an analysis algorithm to identify relevant structures before determining the final optimized reconstruction parameters. This preliminary action enables subsequent optimization without requiring complete re-acquisition of raw data.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The reconstruction parameter set is made dynamic and adaptive rather than static. The system automatically adjusts reconstruction parameters based on analysis information about the specific patient's image content, allowing the parameters to be optimized for each individual case while maintaining high throughput.

Inventive Principle:
Principle #15Dynamics

2Manufacturing precision

If manual re-reconstruction is performed to optimize image quality for individual patients, then image quality improves, but processing time increases and diagnostic workflow is interrupted

Engineering Contradiction:
Improveimage qualityVSAvoidprocessing time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The system performs automatic self-optimization through an analysis algorithm that independently analyzes the preliminary image data set and determines optimal reconstruction parameters without requiring manual intervention from radiologists or technical staff. This eliminates workflow interruptions while maintaining high image quality.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system implements a feedback loop where the analysis algorithm evaluates the preliminary image data set and uses this information to automatically adjust reconstruction parameters for the final medical image data set. This closed-loop feedback enables continuous optimization without manual re-processing.

Inventive Principle:
Principle #23Feedback

3Adaptability or versatility

If multiple reconstruction parameter sets are stored and processed to optimize for different situations, then adaptability improves, but storage space and data transmission requirements increase

Engineering Contradiction:
ImproveadaptabilityVSAvoidstorage space
Core Design Contradiction:
Adaptability or versatilityVSQuantity of substance

Solution Approach 1:

The system performs a preliminary reconstruction with default parameters to generate a preliminary image data set, then uses an analysis algorithm to identify relevant structures before determining the final optimized reconstruction parameters. This preliminary action enables subsequent optimization without requiring complete re-acquisition of raw data.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

Instead of storing multiple complete reconstruction parameter sets for different situations, the system changes parameters dynamically based on analysis information. The analysis algorithm identifies relevant structures and automatically determines appropriate reconstruction parameters, eliminating the need to pre-store multiple parameter sets while maintaining high adaptability.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS20250054140A1Method for determining a medical image data set and provision system
Publication Date: 2025.02.13 SIEMENS HEALTHINEERS AG
  • US20250054140A1 patent drawing
  • US20250054140A1 patent drawing

AI summary

One or more example embodiments relates to a computer-implemented method for determining a medical image data set from raw data using a medical imaging facility. The method comprises reconstructing a preliminary image data set from the raw data via a first reconstruction facility, wherein reconstruction parameters of a predetermined default parameter set are used; analyzing the preliminary image data set via an analysis algorithm on an analysis facility to determine an item of analysis information that describes image content of the preliminary image data set; determining a second parameter set of reconstruction parameters as a function of the analysis information; and determining the medical image data set from at least one of the raw data or the preliminary image data set via the first reconstruction facility or a further reconstruction facility, wherein the reconstruction parameters of the second parameter set are used to determine the medical image data set.