smotlaq / open-watch Goto Github PK

View Code? Open in Web Editor NEW

159.0 4.0 24.0 44.43 MB

An open-source handmade smartwatch. All of the codes, PCBs and schematics are available. ⌚

License: MIT License

C 82.80% Makefile 2.78% HTML 3.02% Assembly 0.18% Java 3.60% CMake 0.01% C++ 7.61%

smart-watch iot stm32 arm hc-05 stm32f030 bluetooth watch pcb max30102

open-watch's Introduction

Table of Contents

About The Project
Donation
Components
Roadmap
Pictures
Results
License
Refereces
Contact

About The Project

Welcome to Open-Watch ✋🏻😃 a wearable device for health monitoring and fitness tracking.

Special thanks to PCBWay for sponsoring us on this open-source project and providing these awesome 4-layer PCBs! Open-Watch is an open-source smartwatch project developed for our B.Sc. final thesis. This smartwatch can measure the linear acceleration of your hand, rotational speed, heart rate, and SpO2 (oxygen saturation). All of these data will be processed by an STM32 processor with an ARM Cortex-M core. You can find these essential components here.

MPU6050 was used for IMU purposes. We use the Kalman filter to reduce the noise effect and extract clean data from the sensor (3-axis linear acceleration, 3-axis rotational velocity, and 3-axis rotation angle).

MAX30102 is the sensor used for health care purposes. We just read raw data and devolve the processes to the smartphone.

Open-watch will send these collected data to a smartphone via a Bluetooth module. On the smartphone side, an Android app reads the data from Bluetooth and processes them. For health care applications like heart rate and SpO2 analysis, we use ML-based algorithms to extract these parameters from raw data read from the pulse-oximeter sensor. In this section a real-time algorithm for analysis of photoplethysmography signal (PPG) for measurement of SpO2 will be implemented.

The final valuable data will be shown on a smartphone as well as the smartwatch screen.

(back to top)

Donation

Do you want to support us in this project?

(back to top)

Components

4-layer PCB: The PCBs of this project are provided by PCBWay, one of the best PCB factories in the world.
Main proccessor: STM32F030C8 (ARM Cortex-M0)
Heartbeat and pulse oximeter sensor: MAX30102
Motion sensor: MPU6050
Display: 1.3" OLED 64*128 pixels
Vibration motor
Bluetooth: HC-05
Touch driver: BS814-A
Battery: 3.7v Li-Po
Battery charger: TP4056

(back to top)

Roadmap

See the open issues for a full list of proposed features (and known issues).

(back to top)

Pictures

OpenWatch

PCB Overview

PCB - 2D

PCB - 3D

PCB - Real

(back to top)

Star history

(back to top)

License

Distributed under the MIT License. See LICENSE.txt for more information.

(back to top)

Refereces

Dataset:

MIMIC-III Database (https://mimic.mit.edu)

[1] Johnson, A., Pollard, T., & Mark, R. (2016). MIMIC-III Clinical Database (version 1.4). PhysioNet. https://doi.org/10.13026/C2XW26.

[2] Moody, B., Moody, G., Villarroel, M., Clifford, G., & Silva, I. (2020). MIMIC-III Waveform Database (version 1.0). PhysioNet. https://doi.org/10.13026/c2607m

[3] Kemp, J., Zhang, K., & Dai, A. (2020). MIMIC-III - SequenceExamples for TensorFlow modeling (version1.0.0). PhysioNet. https://doi.org/10.13026/n2v5-5b32.

Main references:

[1] Kiyasseh et al, CLOCS: Contrastive learning of cardiac signals across space, time, and patients, In Proc. ICML 2021

[2] Torres-Soto, J., Ashley, E.A. Multi-task deep learning for cardiac rhythm detection in wearable devices. npj Digit. Med. 3, 116 (2020). https://doi.org/10.1038/s41746-020-00320-4

[3] El Hajj C, Kyriacou P.A. Cuffless blood pressure estimation from PPG signals and its derivatives using deep learning models, Biomedical Signal Processing and Control. 70 (2021) https://doi.org/10.1016/j.bspc.2021.102984

[4] Solà, J., & Delgado-Gonzalo, R. (2019). The Handbook of Cuffless Blood Pressure Monitoring: A Practical Guide for Clinicians, Researchers, and Engineers. The Handbook of Cuffless Blood Pressure Monitoring. https://doi.org/10.1007/978-3-030-24701-0

(back to top)