- toolkit for full HRV (heart rate variability) analysis of Pulse Sensor data using standard (Linear Regression, Fourier Extrapolation) compared to groundbreaking new techniques (Wavelet transforms, Chaotic Analysis, Neural Networks)
- script for real-time python display of bpm data from Arduino Pulse Sensor, on Arduino Uno, using Matplotlib.
Requirements: Arduino Uno/Arduino IDE or Raspberry Pi, Python 3, Matplotlib, Numpy, PySerial, Arduino Pulse Sensor
- Use NN regression on regressive output from real data (either standard or NN LR) ...?
- Use ALL models on real data
- Build Chaotic model
- Build DeepNNs
-
Fourier Extrapolation
-
Wavelet Extrapolation
-
Chaotic Analysis:
- For HRV:
- http://www.scielo.br/scielo.php?script=sci_arttext&pid=S2358-04292016000500005
- https://www.ncbi.nlm.nih.gov/pubmed/17593181
- http://geoffboeing.com/2015/03/chaos-theory-logistic-map/
- https://www.researchgate.net/publication/306226253_Visual_Analysis_of_Nonlinear_Dynamical_Systems_Chaos_Fractals_Self-Similarity_and_the_Limits_of_Prediction
- For EKG:
- For HRV:
-
Deep NN:
- Deep NN with Wavelet
- HRV and BPV neural network model with wavelet based algorithm calibration: https://pdfs.semanticscholar.org/8e80/4c4fb5efdce51bbdfa5c26930e8a181ddd62.pdf
- Deep NN with Fitzhugh-Nagumo
- Deep neural heart rate variability analysis: https://arxiv.org/pdf/1612.09205.pdf
- Deep NN with Fourier:
- Training Deep Fourier Neural Networks To Fit Time-Series Data: https://arxiv.org/pdf/1405.2262.pdf
- Deep NN with Wavelet
- Linear, Fourier, Wavelet
- Matplotlib
- Chaos
- Pynamical - nonlinear visualization - https://github.com/gboeing/pynamical
- NN LR, Deep NN