Medical Image Annotation Retrieval System

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Radiologists face significant time constraints when comparing current medical images with prior studies to track changes in lesions over time due to the high volume of images and lack of automated systems for linking relevant prior studies, leading to incomplete understanding of temporal progression and interpretation challenges.

Innovation Solution

An annotation support system that determines the context of current medical images, compares them to prior images, and displays context-relevant annotations and images, utilizing a two-pass process to efficiently retrieve and display matching images and annotations based on modality, body part, and region of interest, enabling automatic detection and linkage of relevant prior studies.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If radiologists manually review all prior imaging studies to compare with current images, then comprehensive comparison and accurate diagnosis are improved, but time consumption and workflow efficiency deteriorate

Engineering Contradiction:
Improvediagnostic accuracyVSAvoidtime consumption
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces the manual mechanical process of radiologists reviewing images with an automated computer-based system that uses algorithms to retrieve and compare prior imaging studies. The system automatically extracts features from current and prior images, performs similarity matching, and presents relevant comparisons to the radiologist, eliminating time-consuming manual navigation while maintaining diagnostic accuracy.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces an automated image retrieval and comparison system as an intermediary between the radiologist and the large volume of prior imaging studies. This intermediary automatically processes images, identifies relevant prior studies based on anatomical region and temporal proximity, and presents a curated set of comparisons, reducing the radiologist's workload while ensuring comprehensive review.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Loss of time

If radiologists compare only the most recent prior study to current images, then time consumption is reduced, but understanding of temporal progression and overall change deteriorates

Engineering Contradiction:
Improvetime consumptionVSAvoidtemporal progression information
Core Design Contradiction:
Loss of timeVSLoss of information

Solution Approach 1:

The patent implements a dynamic retrieval strategy that adapts the number of prior studies retrieved based on clinical context. The system can retrieve a variable number of prior studies (not fixed to just one) depending on factors such as the current study's findings, the patient's history, and the need for temporal progression assessment. This allows flexible adjustment between comprehensive review and efficient comparison.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent segments the retrieval process into multiple stages: first retrieving a broad set of prior studies based on basic criteria (anatomical region, modality), then applying more sophisticated filtering and ranking to identify the most relevant comparisons. This segmented approach allows the system to efficiently present a manageable number of high-value comparisons while maintaining the ability to access broader temporal context when needed.

Inventive Principle:
Principle #1Segmentation

3Reliability

If radiologists manually search for prior images with similar views of lesions, then accurate comparison is improved, but time consumption and ease of operation deteriorate

Engineering Contradiction:
Improvecomparison accuracyVSAvoidease of searching
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The patent replaces the manual mechanical process of radiologists visually searching for and comparing images with an automated computer-based system. The system uses image processing algorithms to automatically identify lesions, extract features, and retrieve prior images with similar views and characteristics, eliminating the time-consuming manual search process while maintaining or improving comparison accuracy.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent transforms the search process from manual visual inspection to automated parameter-based matching. The system extracts quantitative parameters from images (such as lesion location, size, shape, and radiological features) and uses these parameters to automatically retrieve and compare relevant prior studies, making the process both more accurate and easier to operate.

Inventive Principle:
Principle #35Parameter changes

4Device complexity

If no automated detection and linkage of prior studies is implemented, then system complexity is reduced, but productivity and workflow efficiency deteriorate

Engineering Contradiction:
Improvesystem complexityVSAvoidworkflow efficiency
Core Design Contradiction:
Device complexityVSProductivity

Solution Approach 1:

The patent replaces manual workflow processes with an automated computer-based system that performs image retrieval, comparison, and analysis. The system automatically links current studies with relevant prior studies based on extracted features and metadata, eliminating manual navigation and comparison tasks while significantly improving workflow efficiency.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent implements a self-service automated system that autonomously performs the tasks of retrieving prior studies, identifying relevant comparisons, and presenting results to the radiologist. The system uses the patient's own imaging data and metadata to automatically configure and execute the retrieval and comparison process without requiring manual setup or intervention.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP2888686B1Automatic detection and retrieval of prior annotations relevant for an imaging study for efficient viewing and reporting
Publication Date: 2021.01.06 KONINKLIJKE PHILIPS NV
  • EP2888686B1 patent drawingFigure 1
  • EP2888686B1 patent drawingFigure 2
  • EP2888686B1 patent drawingFigure 3

AI summary

An annotation support system (10) comprising at least one display device (20). A context extraction module (24, 36) determines a context of a current medical image from a current image study. A matching module (26, 38) compares the context of the current medical image to contexts of prior medical images from prior image studies. A display module (28, 40) displays at least one of context relevant annotations and context relevant medical images from the prior image studies which match the context of the current medical study.