{ "cells": [ { "cell_type": "markdown", "metadata": { "id": "D7qjUvzDS30B" }, "source": [ "# Vascular age - example" ] }, { "cell_type": "markdown", "metadata": { "id": "BjP220de4M8A" }, "source": [ "## Setup" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Import libraries" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "- The 'WFDB' toolbox provides tools for reading data from 'WaveForm DataBase' format (which is the format used on [PhysioNet](https://physionet.org/))\n", "- The 'Neurokit2' toolbox provides tools for physiological signal processing" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "! pip install wfdb\n", "! pip install neurokit2" ] }, { "cell_type": "code", "execution_count": 51, "metadata": { "id": "QKF3h2F0S30K" }, "outputs": [], "source": [ "import neurokit2 as nk2\n", "import wfdb" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
Q: Can you find out more about these packages? What is the 'Waveform DataBase Format', and what tasks could 'Neurokit' be used for?
Hint: See the WFDB documentation, and the the Neurokit documentation.
Q: Do look into PhysioNet and the datasets it hosts.
\n", "Q: Can you find out the details of the 'pleth_1' signal from the database description? How was it measured? Is it ieal for vascular ageing assessment? What alternative signals are available?
\n", "Q: Similarly, can you find out what different activities the data in this database were collected during? See the database description.
\n", "Info: You can find out more about this function here.
\n", "Q: You can find out more about this function here. What frequency content does it exclude from the signal?
\n", "Q: What alternative reference parameters would be ideal?
\n", "Q: Are there any outliers? Which point(s)?
\n", "Q: Can you assess the strength of the correlation?
Hint: Consider using the 'sklearn' package
Q: What other parameters could we extract from the PPG signal? See the tutorials in this book, and also this article.
\n", "