Medical Imaging Analysis Probability Preselection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The existing methods for analyzing medical imaging data sets have a low success rate in identifying abnormalities, leading to inefficient use of specialist time and resources, as many analyses result in irrelevant findings due to the absence of abnormalities.

Innovation Solution

A method that preselects medical imaging data sets based on a probability value, redirecting analysis efforts to those with a lower probability of negative findings, using a control unit with a trained artificial network to automatically assign probability values and prioritize detailed analysis of potentially abnormal images.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If all medical imaging data sets are analyzed in detail by specialists, then no abnormality is missed, but the success rate of identifying abnormalities remains low and specialist time is wasted on normal cases

Engineering Contradiction:
Improvesuccess rate of identifying abnormalitiesVSAvoidspecialist time spent on normal cases
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary analysis using an artificial neural network to assign probability values to medical imaging data sets before specialist review. This preliminary sorting identifies cases with low probability of abnormalities, allowing specialists to skip detailed analysis of these cases while maintaining high detection rates for actual abnormalities.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system enables self-service by automatically filtering and prioritizing medical imaging data sets based on learned patterns from training data. The artificial network independently evaluates and ranks cases, reducing the burden on specialists to manually assess every case while maintaining reliable abnormality detection.

Inventive Principle:
Principle #25Self-service

2Reliability

If detailed analysis is performed on all medical imaging data sets, then comprehensive examination is ensured, but productivity of specialists is reduced due to high volume of normal cases

Engineering Contradiction:
Improvecomprehensive examination coverageVSAvoidanalysis rate of medical imaging data sets
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system performs preliminary classification using probability values assigned by the artificial neural network, separating cases into those requiring detailed analysis and those that can be quickly reviewed or stored without specialist intervention, thereby increasing overall analysis throughput.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system applies different levels of analysis quality to different data sets based on their probability values. High-probability abnormal cases receive comprehensive specialist review, while low-probability normal cases receive automated assessment or minimal review, optimizing resource allocation while maintaining comprehensive coverage.

Inventive Principle:
Principle #3Local quality

3Reliability

If specialists review all medical imaging data sets, then all potential abnormalities are detected, but the effort is inefficiently distributed across cases with varying likelihood of abnormalities

Engineering Contradiction:
Improvedetection of all potential abnormalitiesVSAvoidanalytical effort spent on low-yield cases
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The artificial neural network performs preliminary evaluation of all medical imaging data sets, assigning probability values that indicate the likelihood of abnormalities. This allows specialists to focus their analytical effort on cases with high probability values, maximizing the detection of actual abnormalities while minimizing wasted effort on normal cases.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system changes the parameter of case prioritization from uniform review to probability-based ranking. By using the probability value as a sorting parameter, the system dynamically adjusts which cases receive specialist attention, ensuring that analytical effort is concentrated on the most promising cases while maintaining reliable abnormality detection.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3467770B1Method for analysing a medical imaging data set, system for analysing a medical imaging data set, computer program product and a computer-readable medium
Publication Date: 2022.11.23 SIEMENS HEALTHCARE GMBH
  • EP3467770B1 patent drawingFigure 1
  • EP3467770B1 patent drawingFigure 2

AI summary

Method for analysing a medical imaging data set (11) comprising: - providing the medical imaging data set (11); - assigning a probability value (12) for a negative finding, in particular for a negative finding of a specific type of abnormality, to the medical imaging data set (11), wherein the probability value (12) is based on the image data set (11); and - providing the medical imaging data set (11) automatically either - to an output device (21) for analysing the medical imaging data set (11) or - to a device for storing (20) the medical imaging data set (11) based on the probability value (12).