Radiation Exposure Distribution Prediction Using Trained Functions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing radiation exposure measurement methods, such as DAP chambers, can only provide summed exposure over an irradiated area and not the distribution of exposure, leading to inaccurate estimation of local radiation exposure in imaging systems.
Innovation Solution
A computer-implemented method using trained functions, such as artificial neural networks, to generate exposure distribution data based on setting data, exposure parameters, and intensity distribution data from a radiation detector, allowing for the prediction of exposure distribution without the need for actual measurement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a DAP chamber is used to measure radiation exposure, then the summed exposure over the irradiated area can be obtained, but the distribution of exposure over the area cannot be determined
Solution Approach 1:
The patent segments the irradiated area into multiple pixel elements, with each pixel representing a local exposure region. By measuring or calculating exposure for each pixel individually rather than summing over the entire area, the system preserves spatial distribution information while maintaining overall exposure data.
Solution Approach 2:
The patent transitions from one-dimensional summed exposure measurement (DAP) to two-dimensional exposure distribution mapping. By representing exposure as a matrix of pixel values with spatial coordinates, the system adds a spatial dimension to the measurement, enabling visualization and analysis of exposure patterns across the irradiated area.
2Ease of operation
If homogeneous exposure distribution is assumed based on DAP, then calculation is simplified, but local exposure peaks are underestimated
Solution Approach 1:
The patent applies local quality by allowing different regions (pixels) of the irradiated area to have different exposure values. Instead of assuming uniform exposure across the entire area, the system assigns specific exposure values to each pixel based on the intensity distribution, capturing local variations and peaks in radiation exposure.
Solution Approach 2:
The patent uses the intensity distribution pattern from the radiation detector as a template or copy to generate the exposure distribution. By copying the spatial pattern of radiation intensity and scaling it according to the DAP measurement, the system creates a realistic exposure distribution map that reflects actual radiation patterns without requiring direct measurement at each pixel location.
3Device complexity
If trained functions are used to generate exposure distribution data, then measurement complexity is reduced, but computational processing is required
Solution Approach 1:
The patent replaces complex physical measurement systems (such as arrays of detectors or film) with a computational approach using trained functions. Instead of using mechanical or physical devices to measure exposure at each pixel location, the system uses machine learning models that process intensity distribution data and DAP measurements to generate exposure distribution maps, significantly reducing hardware complexity.
Solution Approach 2:
The trained function acts as an intermediary between the measured quantities (intensity distribution and DAP) and the desired output (exposure distribution). This computational mediator translates the input data into the exposure map, bridging the gap between what is directly measured and what is clinically needed without requiring direct physical measurement of each pixel's exposure.
Data Source
AI summary
A computer-implemented method for more accurately providing exposure distribution data for radiation from a radiation source of an imaging system, the method comprising receiving setting data of the imaging system, receiving a radiation exposure parameter associated with the setting data, receiving data associated with the setting data and relating to an intensity distribution of the radiation on a radiation detector of the imaging system, applying a first trained function to the setting data, the exposure parameter and the intensity distribution data as first input data for the first trained function, the exposure distribution data being generated as first output data from the first trained function, providing the exposure distribution data; the intensity distribution data being provided by a second trained function being applied to the setting data as second input data for the second trained function, and the intensity distribution data being generated as second output data from the second trained function.