Diagnostic Imaging Catheter Tomography With AI Object Annotation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Interpreting tomographic images captured by diagnostic imaging catheters requires specialized skills, making the catheter systems difficult to use for users without adequate training.

Innovation Solution

A program and information processing system that utilize a first model to output types and ranges of objects in tomographic images, assisted by a control unit to generate object arrangement images and display determination probabilities and basis markers, enabling easier interpretation of images.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If tomographic images are captured using a diagnostic imaging catheter, then diagnostic information is obtained, but interpretation requires specialized skills making the system difficult to use

Engineering Contradiction:

Improvediagnostic accuracyVSAvoidease of use

Core Design Contradiction:

ReliabilityVSEase of operation

Solution Approach 1:

A machine learning model serves as an intermediary between the raw tomographic image and the user, automatically identifying and annotating objects within the image. The model processes the complex medical imaging data and presents simplified, annotated results that are easier for users to interpret, thereby maintaining diagnostic accuracy while improving ease of use.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If training is provided to enable proper use of the catheter system, then interpretation accuracy improves, but training time and cost increase

Engineering Contradiction:

Improveinterpretation accuracyVSAvoidtraining time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system provides automated object identification and annotation through machine learning, enabling users to obtain accurate interpretation results without requiring extensive specialized training. The technology serves itself by automatically performing the complex interpretation task that would otherwise require trained professionals, thereby reducing training time and costs while maintaining interpretation accuracy.

Inventive Principle:

Principle #25Self-service

3Measurement precision

If detailed analysis of tomographic images is performed, then diagnostic precision improves, but processing time increases

Engineering Contradiction:

Improvediagnostic precisionVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

Manual mechanical analysis of tomographic images by trained professionals is replaced with automated machine learning algorithms. The AI model rapidly processes the imaging data to identify and annotate objects, achieving high diagnostic precision through sophisticated pattern recognition while significantly reducing the time required compared to manual analysis.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS12444170B2Program, information processing method, method for generating learning model, method for relearning learning model, and information processing system

Publication Date: 2025.10.14 TERUMO KK

AI summary

A program and the like that make a catheter system relatively easy to use. The program including a non-transitory computer-readable medium (CRM) storing computer program code executed by a computer processor that executes a process comprising: acquiring a tomographic image generated using a diagnostic imaging catheter inserted into a lumen organ; and inputting the acquired tomographic image to a first model configured to output types of a plurality of objects included in the tomographic image and ranges of the respective objects in association with each other when the tomographic image is input, and outputting the types and ranges of the objects output from the first model.