Ultrasonic Diagnostic Apparatus Region of Interest Inference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing ultrasonic diagnostic apparatuses require a long time to set a region of interest due to the need for repeated scanning of steered ultrasonic waves before calculating combined blood flow data, which hampers efficient blood flow or lesion inspection.
Innovation Solution
An ultrasonic diagnostic apparatus that employs a trained model, such as a neural network, to quickly set a region of interest on a medical image by identifying blood flow regions using previously set region of interest data as teaching information, allowing for rapid generation and display of blood flow images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If steered ultrasonic waves are repeatedly scanned to calculate combined blood flow data for setting a region of interest, then the accuracy of blood flow inspection is improved, but the time required to set the region of interest increases significantly
Solution Approach 1:
The system performs preliminary scanning to acquire blood flow data before the actual inspection, pre-calculating combined blood flow data and determining regions of interest in advance. This preliminary action stores the processed information for rapid retrieval during subsequent inspections, eliminating the need to repeat time-consuming scanning and processing steps.
Solution Approach 2:
The system creates a copy of the blood flow information and region of interest data from preliminary scans, storing this copied data for reuse. Instead of重新 acquiring and processing data during each inspection, the system retrieves and utilizes the copied information, significantly reducing processing time while maintaining accuracy.
2Reliability
If repeated scanning is performed to obtain combined blood flow data, then the reliability of the blood flow region identification is improved, but the productivity of the inspection process deteriorates
Solution Approach 1:
The system performs preliminary scanning and processing to reliably identify blood flow regions and set regions of interest before the actual inspection begins. This preliminary action ensures high reliability of identification while separating the time-consuming processing from the productive inspection phase, thereby improving overall productivity.
Solution Approach 2:
The system automatically performs the scanning, data combination, and region of interest setting processes without requiring manual intervention during the inspection. The automated self-service approach maintains reliable identification through consistent algorithmic processing while maximizing productivity by eliminating manual operations and reducing inspection time.
3Manufacturing precision
If combined blood flow data is calculated through repeated scanning, then the quality of the ultrasonic image analysis is improved, but the frame rate of the ultrasonic and blood flow images decreases
Solution Approach 1:
The system performs preliminary scanning and calculates combined blood flow data before generating the final ultrasonic and blood flow images. By completing the computationally intensive data combination and analysis in advance, the system ensures high quality image analysis while maintaining high frame rates during actual image generation and display, as the preliminary processed data is readily available for rapid rendering.
Data Source
AI summary
An ultrasonic diagnostic apparatus, a medical imaging apparatus, a training device, an ultrasonic image display method, and a storage medium storing a program that can quickly set a region of interest on an ultrasonic image are provided. The ultrasonic diagnostic apparatus includes an ultrasonic image generation unit configured to generate an ultrasonic image of a subject, an inference unit configured to set, by using a trained model trained with a region of interest set on the ultrasonic image as teaching data, a region of interest on a new generated ultrasonic image, and a display unit configured to display the region of interest set by the inference unit along with the new generated ultrasonic image.