X-Ray Acquisition Parameter Setting with Local Image Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in clinical settings is the lack of standardized software modules and unreliable information transfer between PACS and X-ray imaging systems, making it difficult to utilize previous X-ray acquisition parameters for consistent image quality, especially in mobile or intensive care unit scenarios where wireless internet is unreliable.
Innovation Solution
A computer-implemented method using a local database and machine learning algorithm to identify similar X-ray images and their associated acquisition parameters, allowing radiographers to select the best parameters for an intended imaging task, with the option to make adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a PACS database is used to access previous X-ray images and acquisition parameters, then image quality consistency can be improved, but system reliability deteriorates due to unreliable wireless internet connections in mobile or intensive care unit settings
Solution Approach 1:
The patent creates a local copy of the PACS database on the imaging system's storage device. This local database stores previously obtained X-ray images and their associated acquisition parameters, enabling the system to access and utilize historical data without requiring continuous reliable internet connectivity to a remote PACS server.
Solution Approach 2:
The system performs preliminary actions by pre-loading and storing previous X-ray images and acquisition parameters in the local database before they are needed for new imaging tasks. This advance preparation ensures that when a new imaging task requires reference to previous studies, the data is already available locally without requiring real-time network access.
2Adaptability or versatility
If manual selection of acquisition parameters is used, then flexibility and adaptability are maintained, but subjectivity and variability increase due to radiographer preferences and experience levels
Solution Approach 1:
The system implements feedback by automatically retrieving and presenting previously used acquisition parameters along with their associated X-ray images to the radiographer. This feedback mechanism allows radiographers to review what parameters were successfully used in previous similar cases, enabling more consistent and accurate parameter selection while maintaining the ability to make necessary adjustments.
Solution Approach 2:
The system provides self-service by automatically searching the local database for relevant previous X-ray images and acquisition parameters based on the current imaging task requirements. This automated retrieval process reduces the manual search effort required by radiographers and ensures that historically successful parameters are readily available for reference.
3Ease of operation
If previous X-ray images are seldom available for a specific patient, then the system remains simple to operate, but the utility of having standardized parameter settings is reduced
Solution Approach 1:
The system dynamically adapts its behavior based on availability of previous patient data. When previous X-ray images and acquisition parameters are available in the local database, the system automatically retrieves and presents them for reference. When such historical data is not available, the system gracefully degrades to allowing manual parameter selection without requiring additional user input or complex operations, thus maintaining simplicity while maximizing productivity where data is available.
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
The invention describes a computer-implemented method of setting acquisition parameters (AP) for an intended X-ray imaging task (34), which method comprises: providing a local database (20) with information regarding X-ray images and associated acquisition parameters; obtaining a preliminary image (15); inputting the preliminary image (15) to a machine learning algorithm (21) previously trained to identify one or more X-ray images (16X) in the image database (20) that exhibit similarity to the preliminary image (15); and outputting the acquisition parameters (16AP) of an identified X-ray image (16X) as proposed acquisition parameters (AP) for the X-ray imaging task (34). The invention further describes a data processing system (2) adapted to carry out such a method, and an X-ray imaging apparatus (1) comprising such a data processing system (2).