Ultrasound Apparatus E-wave A-wave Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for diagnosing mitral valve regurgitation using pulse Doppler schemes often fail to accurately identify the positions of the E-wave and A-wave in Doppler waveform images due to noise, aliasing, or other image quality issues.
Innovation Solution
An ultrasound diagnosis apparatus equipped with a processing circuit that obtains Doppler waveform data and uses a trained model to detect the positions of the E-wave and A-wave, even in images with noise or aliasing, by employing a multi-layered network trained with diverse Doppler waveform data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional methods (algorithm calculation or human identification) are used to identify E-wave and A-wave positions, then the process is simple and fast, but the identification accuracy deteriorates when the Doppler waveform image contains noise or aliasing
Solution Approach 1:
A trained model serves as an intermediary between the Doppler waveform image and the identification process. The model is trained with diverse training data including noisy and aliased waveforms, enabling it to accurately identify E-wave and A-wave positions even in challenging image conditions where conventional methods fail.
2Measurement precision
If a trained model with multi-layered network is used to detect E-wave and A-wave positions, then the identification accuracy improves in noisy images, but the device complexity and processing time increase
Solution Approach 1:
The model is trained in advance with diverse training data that includes noisy and aliased Doppler waveform images. This preliminary training enables the model to quickly and accurately identify wave positions during actual use without requiring complex real-time processing, thus reducing processing time while maintaining high accuracy.
3Reliability
If a trained model is used to identify wave positions in Doppler waveform images, then the diagnostic accuracy improves, but the device complexity and cost increase
Solution Approach 1:
The trained model acts as an intelligent intermediary that enhances diagnostic accuracy by accurately identifying E-wave and A-wave positions even in noisy or aliased images. While the model adds computational complexity, it significantly improves reliability of diagnosis by overcoming the limitations of conventional identification methods.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The apparatus accurately identifies the positions of the E-wave and A-wave in Doppler waveform images, improving diagnostic accuracy and throughput, even in challenging image conditions.
Implementation Method 1
detect positions of an E-wave and an A-wave in the first Doppler waveform data by using the first Doppler waveform data and a trained model trained with training data that includes at least positions of an E-wave and an A-wave in each of a plurality of pieces of second Doppler waveform data related to left ventricular blood inflows and the plurality of pieces of second Doppler waveform data
Data Source
AI summary
An apparatus according to an embodiment includes a processing circuit. The processing circuit is configured to obtain first Doppler waveform data related to a left ventricular blood inflow. The processing circuit is configured to detect positions of an E-wave and an A-wave in the first Doppler waveform data, by using the first Doppler waveform data and a trained model trained with training data that includes at least positions of an E-wave and an A-wave in each of a plurality of pieces of second Doppler waveform data related to left ventricular blood inflows and the plurality of pieces of second Doppler waveform data.