Respiratory Status Classification via Lung Segmentation and Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current clinical decision-making tools are insufficient in timely recognizing and diagnosing respiratory-related diseases like ARDS, leading to delayed interventions and high mortality rates due to the lack of effective integration of physiological parameters and chest images for accurate diagnosis.
Innovation Solution
A respiratory status classifying method and system that processes original physiological parameters and chest images using machine learning algorithms to generate characteristic features, training classifiers to differentiate between respiratory statuses, thereby assisting in timely and accurate diagnosis and treatment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If clinical decision supporting tools are developed to assist diagnosis of respiratory-related diseases, then the accuracy of respiratory status classification is improved, but the device complexity increases due to integration of multiple physiological parameters and chest images
Solution Approach 1:
The system segments the diagnostic process into distinct modules: physiological parameter collection module, chest image processing module, feature extraction module, and classification module. Each module processes specific data types independently before integrating them in the classification stage, reducing overall system complexity while maintaining high diagnostic accuracy.
Solution Approach 2:
The patent introduces feature extraction as an intermediary layer between raw data (physiological parameters and chest images) and the classification model. This intermediary processes and transforms raw data into meaningful features, simplifying the integration complexity and improving the accuracy of respiratory status classification.
2Reliability
If multiple physiological parameters and chest images are integrated for diagnosis, then the reliability of ARDS detection is improved, but the loss of time increases due to comprehensive data processing requirements
Solution Approach 1:
The system performs preliminary processing of physiological parameters and chest images before the actual diagnosis. Features are extracted and pre-processed in advance, transforming raw data into ready-to-use feature vectors. This preliminary action reduces the time required during the actual diagnostic decision-making process while maintaining comprehensive data analysis for reliable ARDS detection.
3Productivity
If machine learning algorithms are used to process physiological parameters and chest images, then the productivity of diagnosis is improved, but the device complexity increases due to computational requirements
Solution Approach 1:
The patent extracts and processes only the most relevant features from the comprehensive dataset of physiological parameters and chest images. By identifying and focusing on critical diagnostic features rather than processing all available data equally, the system maintains high diagnostic productivity while reducing computational complexity and resource requirements.
Data Source
AI summary
A respiratory status classifying method is for classifying as one of at least two respiratory statuses and includes an original physiological parameter inputting step, an original chest image inputting step, a characteristic physiological parameter generating step, a characteristic chest image generating step, a training step and a classifier generating step. The characteristic chest image generating step includes processing at least a part of the original chest images, segmenting images of a left lung, a right lung and a heart from each of the original chest images that are processed, and enhancing image data of the images being segmented, so as to generate a plurality of characteristic chest images. The training step includes training two respiratory status classifiers using a plurality of characteristic physiological parameters and the characteristic chest images by at least one machine learning algorithm.