Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Let Your Smartphone Hear Your Breath
Our vision is to create smartphone apps that teach users how to relax based on acquired biomarkers (i.e. respiratory activities) and can, therefore, support them with their stress management or dealing with chronic illness.
Keywords: breathing, biofeedback, android, audio data analysis, machine learning
Biofeedback-based breathing exercise has been shown to be an effective and positive training for stress, immune response, metabolic control and quality of life. However, high-cost medical devices or health professions which are the portal to such training are not easily accessible for general populations.
Taking all the advantages of our Smartphones, from high-quality built-in microphone, computational power, mobility, and to well perceive displays. We are interested in how can a smartphone replace the traditional breathing training and perform as a biomarkers receiver, subsequently present professional health-related information to support users.
In this thesis, we would like to investigate the potential of a breathing detection algorithm in real-time by means of a smartphone microphone based on real-world data. The results would then be expected to be the biofeedback that instructs the users to breathe correctly.
Biofeedback-based breathing exercise has been shown to be an effective and positive training for stress, immune response, metabolic control and quality of life. However, high-cost medical devices or health professions which are the portal to such training are not easily accessible for general populations.
Taking all the advantages of our Smartphones, from high-quality built-in microphone, computational power, mobility, and to well perceive displays. We are interested in how can a smartphone replace the traditional breathing training and perform as a biomarkers receiver, subsequently present professional health-related information to support users.
In this thesis, we would like to investigate the potential of a breathing detection algorithm in real-time by means of a smartphone microphone based on real-world data. The results would then be expected to be the biofeedback that instructs the users to breathe correctly.
Your goal is to create a smartphone app that can detect breathing activities in real-time.
The procedure of this Master thesis would be as follows:
1. briefly review the literature on smartphone microphone audio sensing
2. data sources will be provided for exploration
3. develop and evaluate a predictive model that can predict the ground truth from newly collected data
4. construct an app which embeds the developed model
You should bring:
1. Programming skills, Java, C++ and Python, any of your preference
2. Basic skills and interest in acoustic signal processing and machine/deep learning
3. Interest in digital health and hands-on practice
Your goal is to create a smartphone app that can detect breathing activities in real-time.
The procedure of this Master thesis would be as follows:
1. briefly review the literature on smartphone microphone audio sensing 2. data sources will be provided for exploration 3. develop and evaluate a predictive model that can predict the ground truth from newly collected data 4. construct an app which embeds the developed model
You should bring:
1. Programming skills, Java, C++ and Python, any of your preference 2. Basic skills and interest in acoustic signal processing and machine/deep learning 3. Interest in digital health and hands-on practice