Deciphering Brainwaves with the Web Audio API
summary
Claims that browser JavaScript plus the Web Audio API can parse EEG and aid epilepsy diagnosis by converting recordings into audio buffers, applying filters, then converting back; implements an offline pipeline with OfflineAudioContext, experiments with createIIRFilter (a Butterworth port from MATLAB ~200 lines) and finds IIR creation ≈100 ms per filter (≈300 ms per page) versus target 215 ms, so switches to BiquadFilterNode (~10 ms) which matches desktop filtering and surfaces spike‑wave events in noisy seven‑day recordings.
AI-written summary · about
chapters (23)
- 0:00Introduction
- 0:40About epilepsy
- 6:03What is a filter
- 7:18Highpass filter
- 7:39Filter implementation
- 8:06Package implementation
- 8:44Whats coming next
- 9:13Mozilla Developer Network
- 9:33Audio Context
- 10:10Processing EEG Data
- 11:34The Six Steps
- 11:53AudioContext
- 17:10Demo
- 18:04The problem
- 18:44Where can we go
- 19:37Problem
- 19:49Solution
- 20:26Filters
- 20:53Custom Filters
- 21:39Does it work
- 22:09Bipod Filter
- 22:30EEG Data
- 23:57Closing remarks
related talks
53mBrain: The ultimate hack
34mReading Thoughts with JavaScript
47mReactive Brain Waves
31mLearn to control your brain: Brain Computer Interfacing with Python
35mTranslating Brain Electrical Signals into Actions through Models...
46mControlling apps with your mind
26mBrainDrain Using Machine Learning and Brain Waves to Detect Errors in Human
11mConnected brains