Group Biofeedback via Riemannian Distance Matrix
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current biofeedback methods fail to effectively measure interaction among multiple users in real-time, especially in group settings, due to limitations in scalable interaction measures, sensitivity to outliers, and the need for extensive data points, which hinders timely and accurate monitoring and feedback.
Innovation Solution
A method utilizing Riemannian geometry to compute a regularized interaction matrix from biological signals, providing biofeedback based on Riemannian distance, which is robust to autocorrelation and outliers, and adaptable to heterogeneous datasets, enabling real-time or near real-time analysis and multi-modal integration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If extended interaction measures are used to scale to multiple time-series, then the measure can handle group settings, but they require many data points and fail to provide timely measure in online monitoring
Solution Approach 1:
The patent segments the interaction measurement problem into bivariate measures that can be computed independently for pairs of time-series, then combines them through regularization. This allows the system to scale to multiple users while maintaining computational efficiency suitable for real-time monitoring, avoiding the need for extensive data points required by traditional multivariate measures.
Solution Approach 2:
The patent introduces regularization parameters that control the trade-off between using more data points for accuracy and providing timely feedback. By adjusting these parameters, the system can adapt to different online monitoring requirements while maintaining scalability across multiple time-series.
2Adaptability or versatility
If extended interaction measures are used to scale to multiple time-series, then the measure can handle group settings, but they fail to provide accurate measure in the presence of outliers
Solution Approach 1:
The patent introduces regularization as an intermediary mechanism that mediates between the bivariate interaction measures and the final multivariate assessment. This regularization process filters out the influence of outliers while preserving the scalability to handle multiple time-series, thereby maintaining both adaptability and measurement precision.
3Productivity
If bivariate interaction measures are used, then the measure is computationally efficient, but they fail to provide a modular measure that readily scale to any number of time-series
Solution Approach 1:
The patent merges multiple bivariate interaction measures into a unified multivariate assessment through regularization. This combining approach maintains the computational efficiency of bivariate measures while achieving scalability to any number of time-series, effectively resolving the contradiction between productivity and adaptability.
4Measurement precision
If data integration across multiple datasets is performed, then performance in classification and prediction is improved, but the difficulty in combining data from different nature increases
Solution Approach 1:
The patent creates a universal framework using regularization that can handle multiple types of biological signals (EEG, ECG, GSR) uniformly. This multi-functional approach improves performance in classification and prediction while reducing the complexity of combining heterogeneous data, as the same regularization process applies to all signal types.
Data Source
AI summary
A method for providing biofeedback to a group of users, the method being carried out a plurality of times and including acquiring at least one biological signal using at least one biological sensor for each user of the group of users, receiving, by a processor, the at least one biological signal of each user of the group of users, computing, by the processor, a regularized interaction matrix of a dataset including at least part of the received biological signals, computing, by the processor, a Riemannian distance between the regularized interaction matrix and a pre-computed regularized interaction matrix under no interaction, and providing biofeedback to the group of users using sensory means, the biofeedback being based on the computed Riemannian distance.