Privacy-Preserving Video Seizure Detection in Household Lighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing seizure detection devices are obtrusive, uncomfortable, and may stigmatize users, while video-based methods lack practicality due to marker occlusion and require continuous video storage, compromising privacy.
Innovation Solution
A device integrated into everyday objects uses Optical Flow, Independent Component Analysis, and Machine Learning to detect seizures without markers, ensuring privacy by design, and transmits alerts via power-line communication or wireless protocols.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If video-based seizure detection is implemented, then unobtrusive monitoring is achieved, but user privacy is compromised due to continuous video storage requirements
Solution Approach 1:
The patent extracts only the essential motion information from video data by computing Optical Flow vectors, which represent motion patterns without storing actual video frames. This separates the useful diagnostic information (motion patterns indicating seizures) from the privacy-sensitive information (visual appearance), achieving both unobtrusive monitoring and privacy preservation.
Solution Approach 2:
Instead of storing original video frames, the system creates a transformed copy in the form of Optical Flow feature vectors. These vectors capture the dynamic motion characteristics needed for seizure detection while being mathematically transformed representations that do not reveal personal identity or visual information, thus preserving privacy while maintaining detection capability.
2Measurement precision
If marker-based video detection is used, then movement tracking accuracy is improved, but practicality deteriorates due to marker occlusion
Solution Approach 1:
The patent replaces the mechanical marker-based tracking system with an optical flow-based computational approach. Instead of physically attaching markers to the body that can be occluded, the system uses computer vision algorithms to calculate motion vectors directly from video frames, eliminating the need for physical markers and their associated practicality issues while maintaining measurement precision.
3Measurement precision
If wearable seizure detection devices are used, then detection accuracy is improved, but user comfort deteriorates due to continuous device wearing
Solution Approach 1:
The patent introduces video recording as an intermediary medium between the patient and the detection system. Instead of requiring direct contact with wearable sensors on the patient's body, the system uses a remote video camera to capture motion patterns. This intermediary approach maintains detection accuracy through optical flow analysis while completely eliminating the comfort issues associated with continuous wearable device usage.
Data Source
AI summary
The present invention refers to a video-based system for the detection, recognition, registration and/or segmentation of seizures in an unobtrusive and privacy-preserving manner. The device may be discreetly encased within a household object such as a light fixture (100) to ensure unobtrusiveness and is paired with a motion recognition procedure that generates unidentifiable representations of the data. A computational unit (200) receives video footage from a video camera unit (500) and powers a lighting unit with regular (601) and infrared (602) lighting modules, for operation under any lighting conditions. This computational unit (200) also interfaces with a control module (300) to enable manual control of the recording process and a transmission module (400) to securely transmit recorded data or information about recorded data.