fMRI Data Processing Using Segmented Volume Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The processing of measurement data from functional magnetic resonance tomography (fMRI) requires significant storage and computing resources due to the need to load and analyze large datasets, leading to long computing times and inefficient data handling.
Innovation Solution
Implementing a method where calculations for the general linear model are performed on a sequence of data sets, storing intermediate results and updating them volume-by-volume, allowing for efficient storage and processing by discarding each volume data set after calculation, thus reducing storage needs and enabling real-time applications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the complete measurement data set is loaded into main storage for analysis, then the general linear model can be directly calculated for all volume elements, but the main storage requirement becomes very large (several hundred megabytes to gigabytes)
Solution Approach 1:
The patent divides the complete measurement data set into multiple volume data sets that are processed sequentially. Instead of loading all data simultaneously into main storage, the system processes one volume data set at a time, calculating intermediate results that are stored in smaller buffers. This segmentation allows the analysis to be performed with minimal main storage requirements while maintaining the same analytical accuracy as processing the complete data set.
2Measurement precision
If the complete measurement data set is loaded into main storage, then all time series can be collected and analyzed together, but the computing time becomes very long due to the large data range
Solution Approach 1:
The patent segments the measurement data into volume data sets processed in sequential steps. By calculating intermediate results after processing each volume data set and storing them in compact buffer memory, the system avoids the need to collect and process the entire large data set simultaneously. This reduces computing time significantly while maintaining analytical accuracy through the incremental update of intermediate results.
Solution Approach 2:
The patent performs preliminary calculations of intermediate results after processing each volume data set, before the complete measurement is finished. These intermediate results are stored in buffer memory and can be used for preliminary analysis or real-time applications. The final analysis is completed by updating these intermediate results with the remaining volume data sets, eliminating the need to wait until all data is collected.
3Quantity of substance
If volume-by-volume processing with intermediate result storage is implemented, then main storage requirements are reduced and calculation speed increases, but a buffer memory of appropriate size is required
Solution Approach 1:
The patent changes the storage parameters by transferring data from main storage to buffer memory in a structured manner. The buffer memory is organized to hold intermediate results for all volume elements with dimensions matching the subject volume (e.g., 64x64x32 voxels). This parameter optimization allows the system to use minimal buffer memory (a few megabytes) compared to the gigabytes required in main storage, achieving efficient volume-by-volume processing with manageable memory requirements.
Data Source
AI summary
In a method as well as a device to process measurement data that are composed of a number of data sets with a number of independent random samples originating via temporally successive measurements, for a comparison of the time curve of each acquired random sample with the time curve of a model function using the general linear model, the required calculations are implemented data set-by-data set in a series of the data sets originating from the temporal sequence of measurements, and stored as an intermediate result, with the intermediate results of the directly preceding data set being updated with the new calculations. The comparison can be calculated efficiently and quickly, with a saving of storage space.