Medical Image Tumor Detection Around Organ Cavity Boundaries

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods struggle to accurately detect tumors in medical images when cavities are present in the organ, as the partial volume effect can cause erroneous detection of tumor regions.

Innovation Solution

An image processing device that utilizes a detector to extract organ and tumor candidate regions, a generator to create non-organ regions, and a decision unit to determine the overlap ratio between these regions, removing erroneously detected tumor candidates based on their alignment and overlap with non-organ regions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional tumor detection methods are used on medical images, then tumors can be detected in normal organs, but erroneous detection occurs when cavities are present due to partial volume effect

Engineering Contradiction:
Improvetumor detection accuracyVSAvoiddetection reliability in organs with cavities
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent segments the medical image into multiple regions: organ regions, non-organ regions, and boundary regions. By dividing the image processing into distinct segments, the system can apply different processing strategies to each region, thereby improving tumor detection accuracy while avoiding erroneous detections in cavity areas.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts and identifies boundary regions that exhibit partial volume effects, separating them from genuine tumor regions. By taking out the problematic boundary regions for special handling, the system eliminates the source of erroneous detection while preserving true tumor detections.

Inventive Principle:
Principle #2Taking out (Extraction)

2Productivity

If image processing is performed to detect tumor candidates, then detection speed is improved, but false positives increase in regions with partial volume effects

Engineering Contradiction:
Improvetumor detection speedVSAvoidfalse positive rate
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent performs preliminary identification of boundary regions and non-organ regions before final tumor detection. By preparing these region masks in advance, the system can quickly exclude false positive areas during the detection process, maintaining high detection speed while reducing false positives.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces boundary region masks and non-organ region masks as intermediary elements that mediate between the raw image data and the final tumor detection results. These masks act as filters that prevent false positives from reaching the final detection output.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP4379658B1Method for processing medical image
Publication Date: 2026.03.11 FUJITSU LTD
  • EP4379658B1 patent drawingFigure 1A~1C
  • EP4379658B1 patent drawingFigure 2
  • EP4379658B1 patent drawingFigure 3A~3B

AI summary

An image processing method is executed by a computer and the method includes: extracting an organ region representing an organ and a tumor candidate region having a feature for identifying a tumor in the organ from image data obtained by capturing an image of the organ; generating a non-organ region representing a region where the organ is not present using the image data; and removing, from the extracted tumor candidate region, a tumor candidate region being present only at an outer edge portion of the non-organ region in the organ region.