Masimo Rad-8 Pulse Oximeter Data to InfluxDB

A simple Python3 script to read serial data from a Masimo Rad-8 Pulse Oximeter. Samples are extracted at a frequency of 1 sample/second, placed in JSON format and sent to a local InfluxDB instance. This data is visualized using Grafana.

Purpose

My daughter has a trach and is ventilator dependent. My wife and I need to watch her O2 percentages and were only able to do so if we were sitting in front of the pulse oximeter. Now, using InfluxDB and Grafana running on a local server, we can see her specific oxygen and heartrate from anywhere in our home.

Long-term Goal

My daughter has intermittent desats which I do not fully understand. My hypothesis is that since she also has a VP Shunt that it is related to the barometric pressure. This is why I am also monitoring the barometric pressure, indoor temperature & humidity using Bosch's BME280. After approximately 30-days, I will be exporting the pulse-oximeter and weather data to look for correlation between her breathing and other external factors.

Software Dependencies

1. influxdb (Database Server - Named as such if using Raspbian or other Debian derivative)
2. python3-influxdb (Python 3.x - Client)
3. serial Package (Python 3.x - Necessary for serial communication via RS232/DB-9)
4. grafana (Dashboard frontend and server)
5. smbus2 (Python 3.x - Optional for weather data correlation)
6. bme280 (Python 3.x - Optional for weather data correlation)

Hardware Dependencies

1. Masimo Rad-8 Pulse Oximeter (I have not tested other pulse oximeters)
2. Serial to USB (Male RS232/DB-9 to USB)
3. Raspberry Pi Zero W (Or practically any other flavor of computer. This was $9.00)
4. Bosch bme280 Temp, Humidity & Pressure Sensor (Optional if weather data is desired)

Grafana Dashboard for Pulse Oximeter Data w/ Inside Weather

How to Use

These scripts are installed on a Raspberry Pi Zero W running Debian Buster Lite. The Pi is attached to my daughters pulse oximeter which is sometimes mobile. Due to the pulse oximeter and pi being mobile, I energize the pi using a cell phone battery charger. Because the charge is not infinite and I do not want to hook the pi up to a monitor or ssh in every power cycle, I use a keep_going script. The keep_going script has been setup to run every minute as a cronjob. If the keep_going script sees pox_2_influx running in the background, nothing is done and keep_going will exit. If pox_2_influx is not running, then it will be executed. Please change the path w/in graf_keep_going.sh to reflect the location of pox_2_influx.py.

The bme_influx.py file as mentioned above is to capture the weather parameters from a Bosch BME280 sensor also running on the Raspberry Pi Zero W. The sensor is connected to the i2c bus on port 1 and address 0x77. This script does not need or require the keep_going script but is currently setup to run every minute via a cronjob.

The two primary files are:

1. pox_2_influx.py
2. graf_keep_going.sh

Additional weather tracker using BME280:

1. bme_influx.py

Credit Where Credit is Due

I was using a rather unimpressive web site hosted on my local network to review my daughters real-time spO2 readings until I stumbled upon this repo: https://github.com/yahnatan/pulseox

I would like to thank the above individual for their work. I can now trend, decompose and thouroughly nerd out on my daughters time-series, pulse oximeter data.