max30102-tutorial-beaglebone

This repository is unofficial porting of Arduino sample code of MAXRESDEF117#(max30102) HR/SpO2 sensor.

MAXREFDES117#: https://www.maximintegrated.com/jp/design/reference-design-center/system-board/6300.html
Original Arduino codes: https://github.com/MaximIntegratedRefDesTeam/RD117_ARDUINO/

Related Qiita post (Japanese): https://qiita.com/vrn/items/1ac58c61194b23af1d8c

Files

• max30102.py contains a class which has some setup functions and sensor-reading functions
• hrdump.py uses max30102.py and records the values read by the sensor (sample logs areir.log and red.log)
• hrcalc.py provides a function which calcs HR and SpO2 (but experimental)
• makegraph.py can visualize the log data (but it does not work on raspberry pi / beaglebone cli)

Run

Before running, enable i2c interface, install smbus and rpi.gpio, and connect the sensor.
If you use hrcalc.py to get HR and SpO2, numpy is required.

Note that the following pin connection is assumed in the files:

RPi	HR Sensor
3.3V (pin1)	VIN
I2C_SDA1 (pin3; GPIO 2)	SDA
I2C_SCL1 (pin5; GPIO 3)	SCL
- (pin7; GPIO 4)	INT
GND (pin9)	GND

The code requires all 5 pins to work correctly. Please ensure that INT pin is connected.

Record Raw Values

You can initialize the sensor by running the followings.

>>> import max30102
>>> m = max30102.MAX30102() # sensor initialization
>>> red, ir = m.read_sequential() # get LEDs readings

read_sequential() will block until getting 100 values (default; can be changed) for each LED. It uses INT pin for checking the value availability. If the pin is not connected, this will blocks forever (i.e. the script will freeze).

After that, red and ir should have 100 values each.

If you use different pin other than pin 7 for INT, max30102.MAX30102() should be modified (gpio_pin=7 is the default value).
If your HR sensor's address is not 0x57, max30102.MAX30102() should be modified (address=0x57 is the default value).

Calculate HR / SpO2

hrcalc.py has a function calc_hr_and_spo2(ir_data, red_data) -> (hr, hr_valid, spo2, spo2_valid) which works as an approximation of maxim_heart_rate_and_oxygen_saturation in the original Arduino implementation.

The resulting HR & SpO2 may be different between original implementation and this implementation.

# after you load red and ir
>>> import hrcalc
>>> hrcalc.calc_hr_and_spo2(ir[:100], red[:100]) # give 100 values
(136, True, 98.752554, True)
# this shows hr is 136 and it is a valid value, spo2 is 98% and it is a valid value
# this value is produced when using line 10 - 110 of sample logs

Using sample files (ir.log, red.log)

These two files are pre-recorded sensor data. Both has 1000 lines (= 1000 values).

Tesing can be:

>>> import hrcalc
>>> ir = []
>>> with open("ir.log", "r") as f:
...     for line in f:
...         ir.append(int(line))
...
>>> red = []
>>> with open("red.log", "r") as f:
...     for line in f:
...         red.append(int(line))
...
>>> for i in range(37):
...     print(hrcalc.calc_hr_and_spo2(ir[25*i:25*i+100], red[25*i:25*i+100]))
...
(-999, False, -999, False)
(107, True, 99.43662599999999, True)
(88, True, 99.519096, True)
(83, True, 99.855786, True)
(83, True, 99.59255399999999, True)
(125, True, 99.758856, True)
(100, True, 99.43662599999999, True)
(107, True, 99.657, True)
(115, True, 99.848136, True)
(136, True, 99.824664, True)
(125, True, 99.0534, True)
(100, True, 99.8058, True)
(115, True, 99.135144, True)
(93, True, 99.016626, True)
(100, True, 99.771114, True)
(88, True, 99.169194, True)
(100, True, 99.169194, True)
(107, True, 99.373746, True)
(100, True, 99.84405, True)
(125, True, 99.758856, True)
(115, True, 99.758856, True)
(88, True, 99.43662599999999, True)
(68, True, 99.345144, True)
(65, True, 99.373746, True)
(65, True, 99.373746, True)
(78, True, 99.854424, True)
(88, True, 99.712434, True)
(83, True, 99.657, True)
(93, True, 99.345144, True)
(88, True, 99.712434, True)
(88, True, 99.855786, True)
(68, True, 99.848136, True)
(68, True, 99.848136, True)
(42, True, 99.854424, True)
(71, True, 96.23505, True)
(75, True, 98.928594, True)
(83, True, 99.43662599999999, True)

Running Continuously

You may wrap read_sequential() with while infinite loop to get sensor values continuously.

import max30102
import hrcalc

m = max30102.MAX30102()

# 100 samples are read and used for HR/SpO2 calculation in a single loop
while True:
    red, ir = m.read_sequential()
    print(hrcalc.calc_hr_and_spo2(ir, red))

---

Caution: Do not use these files at critical/fatal situations. No guarantee of calculation result.