Machine-Learning Generated Artwork for Biosignal Visualization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic devices struggle to effectively visualize higher-level features such as stress or workload from biosignals, making it difficult for users to interpret and understand complex biosignal data.
Innovation Solution
A data-driven approach that converts biosignal data into machine-learning generated content, using a generative model to create artwork that reflects the user's physiological and mental states, allowing users to associate artwork changes with different states.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If traditional biosignal visualization methods are used, then basic biosignal data can be displayed, but higher-level features such as stress or workload cannot be effectively visualized
Solution Approach 1:
The patent introduces machine learning models as intermediary components that bridge the gap between raw biosignal data and meaningful visual representations. The ML models process complex biosignal patterns and generate artistic visualizations that encode higher-level features like stress and workload, making them accessible to users without requiring complex data processing knowledge.
Solution Approach 2:
The patent replaces traditional mechanical/technical visualization approaches with AI-driven generative models. Instead of using complex charts and graphs that require technical interpretation, the system uses neural networks to generate artistic content that intuitively represents physiological states, substituting technical complexity with creative representation.
2Ease of operation
If complex biosignal data is displayed directly, then complete information is available, but users find it difficult to interpret and understand
Solution Approach 1:
The patent transforms biosignal parameters into different representations through machine learning. The system changes the parameter space from raw signal values to generated artistic properties (color, shape, composition) that preserve information while being intuitively interpretable. This parameter transformation makes the data both complete and accessible.
Solution Approach 2:
The patent creates visual copies or representations of biosignal states through generative art. Instead of displaying raw data directly, the system generates artistic copies that capture the essence of physiological states. These copies serve as intuitive proxies that users can easily interpret while maintaining the underlying information integrity.
3Ease of operation
If machine-learning generated content is used, then intuitive understanding of physiological states is achieved, but computational resources and processing time increase
Solution Approach 1:
The patent applies partial action by using pre-trained machine learning models that have already processed large datasets during training. During actual use, the system only needs to input new biosignal data and receive generated visualizations, rather than performing complex computations from scratch. This partial application of ML reduces real-time computational energy consumption while maintaining high interpretability.
Data Source
AI summary
A method is provided that includes receiving biosignal data measured from a user, encoding the biosignal data into a vector, and generating, using a generative model, an image based on the vector. The generated image is provided for display.