Detection of Artifacts in Ambulatory Electrodermal Activity Data
summary
An automatic approach detects artifacts in ambulatory EDA using EDA together with skin temperature and wrist acceleration as inputs, segmented into five‑second windows and classified with gradient‑boosting ensembles. A dataset from 13 participants with 107.56 hours manually labeled supports leave‑one‑subject and leave‑one‑day‑out evaluation. The model achieves 98% recall for clean/shape artifact classification (42 percentage points above baseline), the data and annotation guidelines are available on request, and an EDA quality index accounts for movement and temperature.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 0:42Background
- 1:23Existing approaches
- 1:59Data collection
- 2:51Data analysis
- 4:07Results
- 4:51Comparison
- 5:17Conclusion
related talks
46mAnalyzing the ElectroCardioGram (ECG)
17mEMPress: Practical Hand Gesture Classification with Wrist-Mounted EMG and Pressure Sensing
28mHow Smart is the ECG? Lessons from Advanced ECG Analytics
20mAdvancing Hand Gesture Recognition with High Resolution Electrical Impedance Tomography
05mA Feature Adaptive Learning Method for High-Density sEMG-Based Gesture Recognition
01h 13mHealth Monitoring with Wireless Sensors and Machine Learning
22mSeeing, Sensing and Recognizing Laban Movement Qualities
06mOpiTrack: A Wearable-based Clinical Opioid Use Tracker with Temporal Convolutional Attention ...