This repository provides Python code to download measurements via Bluetooth from Microlife's blood pressure monitor BP3GY1-2N which is sold at Costco stores. USB devices, in particular Microlife model BP3GX1-5X, purchased at Costco in February 2015, are also supported.
This code has been developed on Ubuntu 18.04 Linux. It updates the date and time and patient id of the blood pressure monitor and downloads the blood pressure monitor's id and measurements. It inserts the id and measurements into a database.
A version with a GUI - bpm_gui.py - is provided as well. It lists blood pressure and pulse data from the database in a table and plots them. It allows to download data from the blood pressure monitor. It looks like this:
Bluetooth communication of the blood pressure monitor is activated by pressing Start for 8 seconds. When Bluetooth is activated, the Bluetooth logo flashes and the display shows its MAC address. Bluetooth communication of the blood pressure monitor needs to be activated in order for it to be found by this and any other software. Before this code is called for the first time, the operating system's Bluetooth tool should be used to pair the blood pressure monitor with the computer .
This code has been written for BP3GY1-2N, a Bluetooth blood pressure monitor by Microlife that is sold exclusively at Costco. There is a good chance that virtually identical products are sold under other names. This tool looks for a Bluetooth device with service UUID 0000fff0-0000-1000-8000-00805f9b34fb. If such a Bluetooth device is found it will be accepted, the device name BP3GY1-2N does not need to match.
This code has been written in Python3. For Bluetooth communication it relies on package bleak. USB communication is based on hid. The GUI additionally needs PyQt5 and matplotlib. All necessary packages can be installed on Ubuntu Linux with these two commands:
sudo apt install python3-pyqt5 python3-matplotlib python3-pip libhidapi-libusb0
sudo pip3 install bleak hid
The Bluetooth code has recently been rewritten to use package bleak for Bluetooth LE. The USB code requires package hid. Thanks to bleak's and hid's availability on Linux, Windows 10 and macOS, this code should run on all three platforms.
Packages bleak and hid as well as PyQt5 and matplotlib for the GUI, can be installed with this command
pip install bleak hid matplotlib PyQt5
where pip3 should be called instead of pip when appropriate to avoid accidentally installing packages for Python 2.
USB communication is based on package hid. The Python module hid requires its native library hidapi to be installed. Please follow these instructions to install hidapi.
- This code has been developed on Ubuntu Linux 18.04 and it works fine on Linux.
- This code now works fine on Windows 10 as well. The recent ports to
bleakandhidenabled these tools to run on WIndows 10 as well. - No feedback for macOS has been received yet.
Care must be taken when the same computer - or just the same Bluetooth USB dongle - is used with different operating systems. The blood pressure monitor, once paired, does not appear to pair again with the same Bluetooth adapter (to be more precise: with an adapter that has the same mac address but lacks the encryption keys that were negotiated during the pairing). A workaround is to change the adapter's mac address to another one that the blood pressure monitor has not been paired with. Changing the mac address has worked for me with Linux utility bdaddr. A much superior solution is to copy the encryption keys from one operating system to the other. Windows 10 stores them in the registry, and Linux stores them in the file system under directory /var/lib/bluetooth. Since the blood pressure monitor cannot easily be paired again, it is a good idea to keep a backup of the encryption keys.
After Python3 and the dependencies have been installed, this tool can be started by clicking on bpm_gui.py.
The graphical user interface allows to add, edit and delete patient records. It allows to import data written by Microlife's tools. It also communicates via Bluetooth LE and USB with the blood pressure monitor and downloads measurements. Moreover, it sets the date and time and assigns patient ids of the blood pressure monitor.
Besides downloading measurements from the blood pressure monitor, data from Microlife software can be imported as well.
The Microlife software on Windows allows to export blood pressure measurements to .csv files. These files can be imported into our database.
On Android, the Microlife Connect app stores blood pressure measurements in a sqlite database. When the phone is connected to a computer, that database can be copied from file sdcard/DBMLBPA6 to the computer. This tool allows to import the app's database into our database.
Directory import_examples contains a .csv file and a database from an Android phone as example data.
This code uses a sqlite database. This database is found in ~/.local/share/bpm/bpm.db and direct access is possible as this:
sqlite3 ~/.local/share/bpm/bpm.db
.dump
.quit
On Windows, the database is at \Users\USERNAME\AppData\Local\BPM\bpm.db.
If the GUI is not wanted, the command-line tools bpm_bt.py and bpm_usb.py can be used instead. These tools have identical functionality and are based on Bluetooth and USB respectively. We only describe bpm_bt.py here.
After blood pressure measurements have been taken, they can be downloaded with the command-line tool bpm_bt.py. Again, Bluetooth needs to be activated by pressing Start for 8 seconds. When the Bluetooth logo flashes and the display changes to its MAC address, the blood pressure monitor is ready to connect via Bluetooth. bpm_bt.py will find it and connect to it. The run should look like this:
Contacting 3 devices: DA:9A:85:00:A3:52, 51:13:01:D5:02:C4, 7F:FB:7A:6A:CE:46.
Found Bluetooth Blood Pressure Monitor: address DA:9A:85:00:A3:52, name BP3GY1-2N.
2020-05-16 16:29 sys 130 mmHg, dia 81 mmHg, pulse 90 /min
2020-05-17 10:43 sys 133 mmHg, dia 84 mmHg, pulse 58 /min
2020-05-19 19:26 sys 125 mmHg, dia 80 mmHg, pulse 81 /min
2020-05-26 20:53 sys 124 mmHg, dia 79 mmHg, pulse 77 /min
2020-05-27 20:59 sys 126 mmHg, dia 79 mmHg, pulse 71 /min
2020-05-28 10:07 sys 128 mmHg, dia 79 mmHg, pulse 67 /min
2020-05-28 16:37 sys 126 mmHg, dia 80 mmHg, pulse 72 /min
Blood Pressure Monitor communication done, waiting for 2 other Bluetooth LE devices to disconnect: 51:13:01:D5:02:C4, 7F:FB:7A:6A:CE:46.
Downloaded 7 measurements, inserted 3 into database /home/user/.local/share/bpm/bpm.db.
The first line
Contacting 3 device: DA:9A:85:00:A3:52, 51:13:01:D5:02:C4, 7F:FB:7A:6A:CE:46.
lists one or more Bluetooth LE devices that were found during the discovery phase. This software is going to communicate with all of them in order to find the blood pressure monitor.
The line
Found Bluetooth Blood Pressure Monitor: address DA:9A:85:00:A3:52, name BP3GY1-2N.
indicates that the device's Bluetooth LE service and characteristics have matched the ones of the blood pressure monitor. This message is immediately followed by data download from the blood pressure monitor.
The blood pressure measurements are self-explanatory and the final line
Downloaded 7 measurements, inserted 3 into database /home/user/.local/share/bpm/bpm.db.
indicates that of the 7 measurements only 3 were inserted into the database. This happens when from previous runs of the tool, 4 of the 7 measurements are already in the database.
The tools bpm_bt.py and bpm_usb.py are commonly called without arguments. Without arguments they download blood pressure measurements and insert them into the database, the most common usage scenario. Several command-line arguments are supported.
Option --id is available to store a patient id in the blood pressure monitor. This option is required when no patient id is stored in the monitor, e.g. upon the first use. Option --id can also be used to assign a new patient id to the blood pressure monitor.
Another option is --import_csv which is followed by a file name. The Microlife software on Windows allows to export blood pressure measurements to .csv files. This option can be used to import that data into the database.
On Android, the Microlife Connect app stores blood pressure measurements in a sqlite database. When the phone is connected to a computer, that database can be copied from file sdcard/DBMLBPA6 to the computer. This tool's option --import_db allows to import the app's database into our database.
The Microlife Connect app on Android communicates via Bluetooth LE and writes extensive debug logfiles under directory sdcard/Downloads. These logs include all the data that pass back and forth between the app and the blood pressure monitor. The Bluetooth communication code is based on the traffic seen in these debug logs.
For USB, phako has documented and implemented the communication of a Microlife blood pressure monitor. That was extremely helpful. Another implementation is frankyn's.