Brain-Function Image Augmentation via Parameter Space
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current nuclear medicine brain-function image data augmentation methods are inadequate due to subjective image interpretation, lack of standard procedures, and limited data availability, leading to inconsistent and unreliable diagnoses, especially in dementia diagnosis where data imbalance hampers deep learning models.
Innovation Solution
A brain-function image data augmentation method that involves creating a target database with nuclear-medicine images for different ages, calculating image-data ratios, establishing relationships using linear regression, and linearly combining estimated and ratio information to generate augmented data, addressing data shortages and improving deep learning model training.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If rigid body motions (rotations, displacements, scaling) are applied to augment nuclear-medicine brain-function image data, then the number of training samples increases, but physiological characteristics are lost
Solution Approach 1:
The patent changes the augmentation approach from spatial transformations to parameter-space transformations. Instead of rotating or scaling images, it modifies the statistical parameters (mean and standard deviation) of the image data to generate augmented samples that preserve physiological characteristics while increasing data diversity for training.
2Quantity of substance
If more nuclear medical images are collected to improve deep learning training, then data availability increases, but collection costs and time increase significantly
Solution Approach 1:
The patent creates copies of existing images through parameter-space transformations. By generating augmented samples from limited original images using statistical parameter modifications, it produces sufficient training data without the need to collect additional medical images, thus saving time and resources.
3Adaptability or versatility
If subjective image interpretation by physicians is used, then clinical experience can be applied, but reproducibility and consistency decrease
Solution Approach 1:
The patent replaces the subjective human interpretation mechanism with an objective computational mechanism. By using standardized parameter-space augmentation and deep learning models trained on augmented data, it eliminates variability in human judgment while maintaining adaptability through learned patterns, thereby improving reproducibility and consistency.
Data Source
AI summary
A brain-function image data augmentation method includes: a step of providing a target database including a plurality of image-data information; a step of, based on a plurality of image-data expected information in an expectation-value database, calculating a ratio of the plurality of image-data expected information with respect to different ages; a step of, based on the plurality of image-data information and the ratio, obtaining image-data ratio information with respect to an estimated age; a step of establishing a relationship for each pair of the image-data information and the image-data expected information; a step of, based on the relationship, the ratio and the image-data information, calculating an estimated image-data information with respect to the estimated age; and, a step of combining linearly the estimated image-data information and the image-data ratio information so as to generate an augmented image-data information with respect to the estimated age.