Analyzing the ElectroCardioGram (ECG)
summary
Home-recorded three-lead ECG (V2) loads from disk with h5py, or from Physionet via python-wfdb, then a Butterworth bandpass and smoothing reduce noise, wavelet denoising via pywavelets isolates sharp features, and peakutils or CWT finds R peaks with claimed accuracies near 99…; RR intervals give heart rate and HRV (reference ~170 ms); pipeline yields numeric features for ML; demonstration runs in an IPython notebook and the processing and peak detection reliably work on the shown examples.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 3:09Home ECG recording
- 6:16Filtering ECG signals
- 9:01Wavelet transformations
- 13:13ECG peak detection
- 18:53Heart rate calculations
- 25:37Heart rate variability
- 28:45Machine learning in ECG
- 34:04Q&A: Pre-processing
- 38:52Q&A: HRV and peaks
- 43:23Q&A: ML in pharma
- 45:30Closing
related talks
28mHow Smart is the ECG? Lessons from Advanced ECG Analytics
06mHeartQuake: Accurate Low-Cost Non-Invasive ECG Monitoring Using Bed-Mounted Geophones
07mCardiacWave: A mmWave-based Scheme of Non-Contact and High-Definition Heart Activity Computing
06mDetection of Artifacts in Ambulatory Electrodermal Activity Data
27m"Boosting Neuroimaging Analysis and Results Digestibility 🧠"
45mAnomaly detection with TensorFlow
29mMaking Neuroimage Analysis Easier to Understand
05mRF Vital Sign Sensing Under Free Body Movement