Automated Extraction of Annotations from Medical Images

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Manual entry of annotations and alphanumeric information from medical images into clinical reports is tedious and time-consuming, as this information is often not readily accessible within the reporting system, requiring clinicians to view images, remember details, and manually input data.

Innovation Solution

A viewing/reporting system that extracts graphical and alphanumeric information from medical images, including annotations and measurement data, and automatically inserts this information into an electronically formatted clinical report, utilizing a processor with an information extraction module and report generating/editing module to evaluate pixels and integrate the data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual entry of annotations and alphanumeric information is used, then clinicians can accurately document image data, but the process becomes tedious and time-consuming

Engineering Contradiction:

Improveaccuracy of documentationVSAvoidtime required for data entry

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system enables self-service by automatically extracting annotations and alphanumeric information from medical images and populating clinical reports without requiring clinician intervention. The extraction module identifies and captures measurement data, annotations, and metadata, while the report generator automatically integrates this information into the final report, allowing the system to serve itself rather than requiring manual clinician input

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces the mechanical manual entry process with an automated computational system. The extraction module uses image processing and pattern recognition algorithms to identify and extract graphical and alphanumeric information, substituting the manual mechanical process of reading, remembering, and typing with an automated electronic extraction and integration system

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If manual entry process is used, then clinicians can review and verify information, but the process requires excessive time and effort

Engineering Contradiction:

Improveverification of informationVSAvoidreporting efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The system incorporates feedback mechanisms where the extraction module continuously monitors and identifies annotations and information elements, providing real-time feedback to the report generation process. The system verifies extracted information by cross-referencing multiple data sources and validation rules, ensuring accuracy while maintaining automated high-speed processing

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system performs preliminary extraction and validation of information before final report generation. The extraction module pre-identifies and captures all necessary annotations and alphanumeric data, pre-validates the information against predefined criteria, and pre-integrates it into the report structure, eliminating the need for time-consuming manual review while ensuring accuracy

Inventive Principle:

Principle #10Preliminary action

3Ease of operation

If information is not readily accessible in the reporting system, then manual entry is required, but this creates a tedious workflow

Engineering Contradiction:

Improveaccessibility of informationVSAvoidtime for data retrieval and entry

Core Design Contradiction:

Ease of operationVSLoss of time

Solution Approach 1:

The patent merges the image viewing function with the report generation function into an integrated system. The extraction module is embedded within the reporting system, allowing it to directly access and extract information from displayed images without requiring separate data retrieval steps. This merging eliminates the need to switch between systems or manually transfer data, making information readily accessible and eliminating tedious workflow steps

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentEP3100209B1Extraction of information from an image and inclusion thereof in a clinical report

Publication Date: 2022.11.02 KONINKLIJKE PHILIPS NV

Figure 1
Figure 2
Figure 3~4

AI summary

A method includes obtaining, in electronic format, an image (102) including a medical image display region (104) and an information display region (106). The at least one of the medical image display region or the information display region includes graphical indicia representing at least one of an annotation (110, 112, 114, 116) or alphanumeric information (118, 120, 122, 124, 126, 128, 130, 132, 134, 136, 138, 140). The method further includes evaluating pixels of the image to identify pixels including the graphical indicia representing an annotation or alphanumeric information of interest in the image. The method further includes extracting the annotation or alphanumeric information of interest from the identified graphical indicia from the image. The method further includes inserting the extracted annotation or alphanumeric information of interest in an electronically formatted clinical report for the image.