USER INTERFACE, COMMUNICATION SUB-SYSTEM DE- SIGN, AND CASING DESIGN OF PPG BASED CONTINOUS WEARABLE HEART RATE MONITORING DEVICE

• Main Author: Janitra Adi Pratama, Rhesa
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/71835
• Institution: Institut Teknologi Bandung

The heart is an organ that plays a major role in the body. Unmaintained and unmonitored heart condition and health can lead to lots of heart diseases. To be able to maintain heart health and condition, it is necessary to regularly and periodically monitor our heart so that heart diseases’ indications can be identi- fied immediately. The conventional method currently used for measuring heart rate is with electrocardiograph (ECG). This method is known to have accurate and precise results, but unfortunately has a complicated procedure that requires electrodes to be placed at specific points on the body which only qualified medical personels can execute the process with medical knowledge and considerations, in certain medical facilities. Because ECG can only be accessed in certain medical facilities, continuous heart rate checks are very limited, and usually only patients who have heart disease indications are allowed to use ECGs, not for regular med- ical checkups. “PPG-based Continuous Heart Rate Monitoring Wearable device” is a tool worn on the wrist, that can be used to monitor heart rate in real-time and continuously. The implementation of wearable devices is carried out on an ESP32 microprocessor using the Arduino IDE software. The data acquired by the weara- ble is PPG data, using DFRobot, as well as x, y, and z accelerometer data, using MPU6050, with a data retrieval frequency of 50Hz to 60Hz. Data processing and display of heart rate data are carried out in the Android application with the help of the Chaquopy module, where the data is sent via Bluetooth communication. Android applications are implemented on Android Studio software for both back- end and front-end programming. Detection of user activity is implemented using machine learning in the Tensorflow software. The results of processing heart rate data with a resolution of 1 BPM can be stored in smartphone memory and the cloud (AWS).