Discriminative Model for Infarction Detection in Non-Contrast CT

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods for diagnosing cerebral infarction using non-contrast CT images face challenges in accurately specifying the infarction region and large vessel occlusion region due to low contrast between regions and difficulty in distinguishing occlusions from similar structures, leading to delayed treatment and poor prognosis.

Innovation Solution

An information processing apparatus and method that utilize a discriminative model trained on non-contrast CT images to derive information about infarction and large vessel occlusion regions by combining anatomical and clinical information, with the model using symmetrical brain regions and inversion techniques to enhance accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If non-contrast CT image is used for initial diagnosis, then treatment time is reduced, but measurement precision of infarction region and large vessel occlusion region deteriorates

Engineering Contradiction:

Improvetreatment timeVSAvoidspecification accuracy of infarction region and large vessel occlusion region

Core Design Contradiction:

Loss of timeVSMeasurement precision

Solution Approach 1:

A discriminative model serves as an intermediary between the non-contrast CT image and the diagnostic information. The model processes the low-contrast image data and outputs enhanced diagnostic information about infarction regions and large vessel occlusions, enabling accurate diagnosis without requiring contrast agents or additional imaging modalities

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the mechanical/physical approach of using contrast agents and multiple imaging modalities (CT, MRI) with an information processing approach using a discriminative model. The model substitutes for the need for physical contrast enhancement and multi-modal imaging by computationally extracting diagnostic information from non-contrast CT images

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If MRI image or contrast CT image is acquired after non-contrast CT diagnosis, then measurement precision of infarction region and large vessel occlusion region is improved, but loss of time increases

Engineering Contradiction:

Improvespecification accuracy of infarction region and large vessel occlusion regionVSAvoidelapsed time from infarction development

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The discriminative model performs preliminary extraction of diagnostic information from the initial non-contrast CT image before additional imaging is acquired. By pre-processing the available data to extract infarction and occlusion information, the system eliminates or reduces the need for time-consuming follow-up MRI or contrast CT scans

Inventive Principle:

Principle #10Preliminary action

3Measurement precision

If discriminator is trained to extract infarction region and thrombus region from non-contrast CT image, then measurement precision is improved, but device complexity increases

Engineering Contradiction:

Improveextraction accuracy of infarction region and thrombus regionVSAvoidcomplexity of discriminative model training and processing

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The discriminative model is designed to perform multiple diagnostic functions simultaneously: extracting infarction regions, identifying large vessel occlusions, and providing diagnostic information all from a single non-contrast CT image input. This multi-functionality reduces the need for separate specialized systems for each diagnostic task

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS20240415473A1Information processing apparatus, information processing method, information processing program, learning device, learning method, learning program, and discriminative model

Publication Date: 2024.12.19 FUJIFILM CORP

AI summary

A processor acquires at least one of first information representing any one of an infarction region or a large vessel occlusion region in a non-contrast CT image of a head of a patient, information representing an anatomical region of a brain, or clinical information, acquires second information representing a candidate of the other of the infarction region or the large vessel occlusion region in the non-contrast CT image, and derives third information representing the other of the infarction region or the large vessel occlusion region in the non-contrast CT image based on at least one of the first information, the information representing the anatomical region of the brain, or the clinical information, and the second information.