AI DLCO Prediction from Spirometry Flow-Volume Curve Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing pulmonary function tests, particularly the diffusing capacity test for carbon monoxide (DLCO), are cumbersome and challenging for patients, leading to inconsistent results and discomfort, especially for those with pulmonary insufficiency, such as idiopathic pulmonary fibrosis (IPF).
Innovation Solution
A diffusing capacity predicting apparatus using a flow-volume curve image, employing artificial intelligence through a first machine learning model to extract features from spirometry data and a second model to predict DLCO values, without the need for traditional diffusing capacity tests.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the diffusing capacity test is performed to accurately measure gas exchange capacity, then measurement precision is improved, but device complexity and ease of operation deteriorate due to the complex procedure requiring inhalation, breath-holding, and exhalation
Solution Approach 1:
The patent uses an artificial intelligence model as an intermediary to translate simple spirometry flow-volume curve data into accurate diffusing capacity predictions. The AI model acts as a mediator that converts easily obtainable spirometry measurements into clinically valuable diffusing capacity information without requiring complex test procedures
Solution Approach 2:
The patent replaces the mechanical/physiological complexity of the diffusing capacity test (inhalation, breath-holding, exhalation coordination) with a computational system that processes simple flow-volume curve data. The mechanical testing procedure is substituted with an information processing system that automatically extracts and analyzes relevant features
2Measurement precision
If the diffusing capacity test is performed to evaluate lung function, then measurement precision is improved, but loss of time increases due to the prolonged test duration required for proper execution
Solution Approach 1:
The patent extracts the essential measurement information needed for diffusing capacity assessment from the simpler spirometry flow-volume curve. By taking out and analyzing the relevant features already present in routine spirometry data, the system eliminates the need for separate, time-consuming diffusing capacity testing procedures
3Measurement precision
If the diffusing capacity test is performed to obtain accurate results, then measurement precision is improved, but reliability deteriorates due to inconsistent results from patient difficulty in performing the test properly
Solution Approach 1:
The patent creates a computational model that copies and simulates the complex gas exchange relationships based on simpler flow-volume curve patterns. The AI model learns from training data the relationships between spirometry parameters and diffusing capacity, then applies this learned knowledge to predict diffusing capacity from new spirometry measurements, ensuring consistent and reliable results
Data Source
AI summary
A diffusing capacity predicting apparatus according to the present disclosure includes a memory including one or more instructions and a processor which executes the one or more instructions stored in the memory. In the memory, first information about a spirometry result of a user, second information which is clinical information, a first machine learning model, and a second machine learning model are recorded. The processor inputs the first information to the first machine learning model to extract a feature related to a lung disease and inputs the extracted feature and the second information to the second machine learning model to predict a diffusing capacity value of the user. Accordingly, the diffusing capacity may be more accurately predicted using the spirometer result and the artificial intelligence model, without using the diffusing capacity test method.