Recording Your Every Heart Beat for a Healthier Life

                                     Pulse Monitor

Recording your every heart beat for a healthier life

Faculty of Computer Science, Simon Fraser University

CMPT 340 Biomedical Computing

Instructor: Ghassan Hamarneh

Abstract

Heart rate can be important since it carries many information of a person’s health. Nowadays, there are many ways to measure heart rate. Traditional approach for capturing heart rate needs electro-nodes contacting patients’ skin directly. However, it would be very inconvenient when we want to monitor heart rate all day long. Since most people have a smartphone with high-resolution camera, it will be easy to use mobile phone because it does not require any special skills or buying special devices.

In this paper, we will overview current heart rate detecting algorithm by using mobile phone and propose improved method to make it more efficient.

Keywords

Mobile Health, Pulse detection, Mobile phone, Camera, Video Magnification, Moving Average Filtering.

Introduction

Heart rate is the number of poundings of the heart per unit of time. Normally, it will be presented in the beats per minutes (bpm). Heart rate abnormality can indicate diseases such as acute myocardial infarction and heart attack.

The most widely used way to measure heart rate is manual measuring. However, if people want higher accuracy, some devices such as electrocardiographs and pulse oximeter are needed. Other than being more accurate, the other advantage is that by using devices, heart rate can be measured during exercising. This is especially useful when people want to track their cardio work during different level of exercises.

To use mobile phone for measuring heart rate is straightforward. Users can simply launch the application on the phone and put the fingers on the camera, then wait for the result. Therefore we decided to create an application to achieve this.

There are several existing algorithms to do this on the mobile phone. The main idea of them is to detect variation in finger skin color. However, the performance is not reliable when in the real applications. This is because raw data collected by camera is not effectively filtered and enhanced.

We decide to improve those algorithms by applying two signal-enhancing method: video magnification and moving average filter. By combing two techniques together, the algorithm will be more efficient than existing ones.

Our application is named Pulse Monitor. It is an IOS application that utilizes build in cameras in iPhone devices to track the user's heart rate and analyze it based on personal information. The users can choose to record heart rate during rest, before exercise and after exercise. The results that are recorded are then presented to the users in the form of ECG (electrocardiogram) on screen.