Asthma Condition Estimation Using Symptom-Based Patient Clusters

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for evaluating asthma disease conditions require invasive testing and specialized equipment, making them impractical for routine clinical settings, and do not account for individual patient responsiveness to treatment.

Innovation Solution

An asthmatic disease condition estimation server that classifies patients into clusters based on symptom information using machine learning, allowing for the identification of disease conditions without invasive testing, utilizing a storage to associate clusters with disease conditions and a determination unit to estimate disease states from asthma control questionnaire responses.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If invasive testing (bronchoscopy) and gene expression analysis are used to evaluate asthma disease conditions, then measurement precision is improved, but device complexity and ease of operation deteriorate

Engineering Contradiction:
Improvedisease condition evaluation accuracyVSAvoidtesting equipment requirements
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent creates a virtual model (clone) of the patient's asthma disease condition by training a machine learning model on gene expression data and clinical information from training patients. This virtual model can be copied and applied to evaluate disease conditions without requiring actual invasive testing equipment, thereby maintaining measurement precision while eliminating device complexity

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent replaces the mechanical/invasive testing system (bronchoscopy, physical gene expression analysis equipment) with an information processing system that uses machine learning algorithms to analyze gene expression data and clinical information, substituting physical measurement devices with computational models

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

2Measurement precision

If invasive testing (bronchoscopy) is performed to extract airway epithelial cells, then measurement precision is improved, but object-affected harmful factors increase

Engineering Contradiction:
Improvedisease condition evaluation accuracyVSAvoidrisk of complications
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent creates a virtual representation of the patient's disease state through machine learning models trained on existing data, eliminating the need for invasive bronchoscopy procedures that carry risks of complications, while still achieving accurate disease condition evaluation

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent introduces gene expression data and clinical information as intermediary elements that can be obtained through less invasive means (such as sputum samples or blood tests) and serves as a mediator between the patient and the disease evaluation process, avoiding direct invasive intervention

Inventive Principle:
Principle #24Intermediary (Mediator)

3Measurement precision

If gene expression analysis is performed to classify asthma phenotypes, then measurement precision is improved, but loss of substance increases

Engineering Contradiction:
Improvephenotype classification accuracyVSAvoidcost
Core Design Contradiction:
Measurement precisionVSLoss of substance

Solution Approach 1:

The patent creates a computationally-derived copy of the gene expression profile that can be generated from smaller, cheaper samples (such as sputum or blood) rather than requiring large amounts of tissue from invasive bronchoscopy, thereby maintaining classification accuracy while reducing material loss and cost

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent uses a subset of gene expression data and clinical information that is sufficient for accurate phenotype classification without requiring complete comprehensive analysis, thereby achieving the necessary measurement precision with reduced resource investment

Inventive Principle:
Principle #16Partial or excessive action

4Measurement precision

If specialized testing equipment and expertise are required for asthma evaluation, then measurement precision is improved, but ease of operation deteriorates

Engineering Contradiction:
Improvedisease condition evaluation accuracyVSAvoidclinical setting accessibility
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The patent creates a portable, software-based evaluation system that copies the functionality of complex testing equipment into a machine learning model that can run on standard computers or mobile devices, eliminating the need for specialized hardware and making the system accessible in routine clinical settings

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent develops a universal machine learning model that can process multiple types of input data (gene expression data, clinical information, questionnaire responses) and provide comprehensive disease condition evaluation across different asthma phenotypes, replacing multiple specialized tests with a single multi-functional system

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP4636780A1Asthmatic disease condition estimation server, asthmatic disease condition estimation program, and asthmatic disease condition estimation method
Publication Date: 2025.10.22 YAMAGUCHI UNIV
  • EP4636780A1 patent drawingFigure 1
  • EP4636780A1 patent drawingFigure 2
  • EP4636780A1 patent drawingFigure 3

AI summary

To easily estimate a disease condition of a patient with asthma. An asthmatic disease condition estimation server (1) according to the present invention includes: a storage (2) that stores a plurality of clusters, into which patients with asthma are classified, and a plurality of disease conditions of asthma associated with each cluster; a reception unit (3) that receives symptom information indicating a symptom of a subject; a determination unit (4) that identifies the cluster to which the subject belongs among the plurality of clusters, based on the symptom information, and identifies the disease condition stored in the storage in association with the identified cluster; and an output unit (5) that outputs the identified disease condition as an estimated disease condition of the subject.