Medical Image Interpretation Support Apparatus for Detecting Overlooked Regions

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods for mechanically creating medical image interpretation reports have low reliability compared to those created by radiologists, necessitating an improvement in interpretation techniques that may impact patient care.

Innovation Solution

An interpretation support apparatus and computer-readable medium that compare multiple interpretation reports from the same patient to detect regions of interest and findings, identifying differences and notifying users, while also utilizing a correspondence table to standardize terms and deforming medical images for overlay display.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If mechanical methods are used to create interpretation reports, then productivity is improved, but reliability deteriorates

Engineering Contradiction:

Improveinterpretation report creation efficiencyVSAvoidinterpretation report accuracy

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system automatically compares the current interpretation report with historical reports and provides feedback by notifying the radiologist of any differences or overlooked regions, enabling continuous improvement and verification of interpretation accuracy

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system acts as an intermediary between mechanical interpretation methods and radiologist expertise, using AI to detect regions of interest and compare findings while the radiologist maintains final oversight, combining automated efficiency with human reliability

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If multiple interpretation reports are compared to improve reliability, then measurement precision is improved, but device complexity increases

Engineering Contradiction:

Improveregion of interest detection accuracyVSAvoidsystem processing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system extracts only the essential comparison elements (regions of interest and findings) from multiple interpretation reports, focusing computational resources on critical differences rather than processing entire reports, thereby managing complexity while maintaining precision

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS11238976B2Interpretation support apparatus and non-transitory computer readable medium

Publication Date: 2022.02.01 FUJIFILM CORP

AI summary

A support program is a program for supporting creation of an interpretation report including a medical image obtained by examining a patient and a region of interest to focus on in the medical image. The support program causes a computer to function as: a reading unit that reads a plurality of interpretation reports created by a plurality of examinations on the same part of the same patient; a difference detection unit that sets one of the plurality of interpretation reports as a detection target interpretation report and sets the others as comparison target interpretation reports, compares regions of interest of the detection target interpretation report and the comparison target interpretation reports, and detects a region of interest, which is present in the comparison target interpretation reports but is not present in the detection target interpretation report, as a difference region; and a notification screen distribution unit that, in a case where the difference region is detected, notifies that the difference region has been detected.