Medical Image ROI Correction Using Image Features and User Feedback

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image diagnosis systems face challenges in accurately detecting regions of interest in medical images due to inclusion of extra peripheral ranges or omission of detection, requiring manual correction by users, which is time-consuming and inefficient.

Innovation Solution

An information processing apparatus that utilizes learning models to assist in correcting regions of interest by receiving user input corrections and specifying a second region of interest based on image features and correction instructions, reducing the need for manual adjustment.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual correction of detected regions of interest is performed, then accuracy of region specification is improved, but time consumption and operational effort increase

Engineering Contradiction:
Improveaccuracy of region specificationVSAvoidtime consumption for correction
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary detection of regions of interest using CAD before manual correction, automatically identifying candidate regions that require verification. This preliminary action reduces the overall correction time by pre-processing the image analysis and presenting only relevant regions for manual verification.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system incorporates feedback mechanisms where the correction results from manual verification are fed back to improve the CAD detection algorithm. This feedback loop continuously refines the detection accuracy, reducing the need for extensive manual correction over time while maintaining high precision.

Inventive Principle:
Principle #23Feedback

2Measurement precision

If manual correction of detected regions of interest is performed, then accuracy of region specification is improved, but operational effort increases

Engineering Contradiction:
Improveaccuracy of region specificationVSAvoidoperational effort for correction
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The CAD system performs self-service by automatically detecting and highlighting regions of interest that require verification. This automation reduces operational effort by eliminating the need for operators to manually search and identify all potential regions, focusing their effort only on verifying CAD-detected candidates.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system replaces manual mechanical correction operations with automated CAD detection algorithms. By substituting the mechanical process of manual region drawing and adjustment with intelligent software-based detection, operational effort is significantly reduced while maintaining accuracy through subsequent verification.

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

3Productivity

If CAD detection is used to identify regions of interest, then detection speed is improved, but detection accuracy deteriorates due to extra peripheral ranges or omitted peripheral portions

Engineering Contradiction:
Improvedetection speedVSAvoiddetection accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

CAD detection serves as a preliminary action that quickly identifies candidate regions of interest before final verification. The system uses CAD to pre-process and highlight potential regions, maintaining high detection speed while flagging areas that require subsequent manual verification to ensure accuracy.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements feedback where manual verification results are used to refine and improve the CAD detection algorithm. This continuous feedback loop allows the system to maintain high detection speed while progressively improving detection accuracy by learning from correction patterns and reducing false positives and negatives.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP4343695B1Information processing apparatus, information processing method, and information processing program
Publication Date: 2026.05.06 FUJIFILM CORP
  • EP4343695B1 patent drawingFigure 1~2
  • EP4343695B1 patent drawingFigure 3
  • EP4343695B1 patent drawingFigure 4~5

AI summary

Provided are an information processing apparatus, an information processing method, and an information processing program capable of supporting interpretation of images. An information processing apparatus includes at least one processor, and the processor is configured to: acquire an image; display, on a display, a figure indicating a first region of interest included in the image in a superimposed manner on the image; receive a correction instruction for at least a part of the figure; and specify a second region of interest that at least partially overlaps with the first region of interest based on an image feature of the image and the correction instruction.