Brain Function Data Conversion for Disease Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional techniques for detecting brain diseases at an early stage face challenges in accurately determining brain diseases and identifying disease regions from multidimensional data, particularly those involving temporal changes.
Innovation Solution
A brain function determination apparatus that acquires and converts brain function data to include time and space dimensions, using a deep learning model for identification, enabling accurate determination and visualization of brain diseases and their regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional feature extraction techniques are used for brain disease detection, then the detection process is simple, but the accuracy in determining brain diseases and identifying disease regions from multidimensional temporal data is insufficient
Solution Approach 1:
The patent transforms the input data from conventional feature extraction format to a format that explicitly includes time and space as dimensions. This parameter change enables the deep learning model to process multidimensional temporal data effectively, improving disease detection accuracy while maintaining manageable complexity through structured data organization.
Solution Approach 2:
The patent introduces explicit time and space dimensions to the data structure, converting traditional feature extraction output into multidimensional converted data. This dimensionality change allows the deep learning model to capture temporal changes and spatial relationships in brain function data, significantly improving the accuracy of disease determination and region identification.
2Measurement precision
If deep learning models are used to process multidimensional temporal data, then the accuracy of brain disease determination improves, but the computational complexity and processing time increase
Solution Approach 1:
The patent performs preliminary data conversion before deep learning processing, transforming raw brain function data into a standardized format with explicit time and space dimensions. This preliminary action prepares the data in advance, reducing the computational burden during the deep learning inference stage and thereby decreasing overall processing time while maintaining high accuracy.
3Ease of manufacture
If traditional machine learning classification is used, then the system is easier to implement, but it cannot accurately identify brain disease regions from temporal change data
Solution Approach 1:
The patent adds explicit time and space dimensions to the data structure, enabling the preservation and effective utilization of temporal change information. This dimensionality enhancement allows the system to capture dynamic patterns in brain function data that traditional machine learning would miss, while the structured approach keeps the system relatively easy to implement.
Data Source
AI summary
An aspect of the present invention, a brain function determination apparatus includes a first acquisition unit, a first conversion unit, and an identification unit. The first acquisition unit is configured to acquire brain function data including a temporal change, indicating a brain function state measured by a measurement apparatus. The first conversion unit is configured to convert the brain function data acquired by the first acquisition unit, to first converted data including information on at least a time and a space as dimensions. The identification unit is configured to perform an identification process of determining a brain disease and identifying a brain disease region, using the first converted data as an input of a deep learning model constructed by predetermined deep learning.