Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Bachelor/Master Thesis: Development of a ML algorithm for the detection of sleep quality based on wearable data
Sleep is vital for humans, but hardly anyone knows why they slept well. Meanwhile, wearables like smartwatches or smart rings allow us to passively capture various details about our day and night. In this work, you will use state-of-the-art machine learning approaches to quantify sleep quality.
The improvement of wearables allow to collect a high quality data completely automatically, such as the heart rate, activities throughout the day or sleep times. On the other hand, sleep is an important factor for a healthy and happy life and is strongly determined by what is experienced during the day and at night.
In the context of this thesis you should investigate on the basis of self-assessed sleep phases why sleep is better on some days than on others. Moreover, by using state-of-the-art machine learning methods, you should determine whether algorithms can separate good from bad sleep phases. Specifically, you will deep dive into the following steps:
- Use and extend our existing data about sleep behavior and its self-assessment
- Prepare sensor data from wearables such as Garmin smartwatches or Oura Ring for further pre-processing
- Pre-process the available sensor data as features that may have an influence on sleep based on related literature
- Interpretation of the calculated features on the sleep quality with machine learning models and explainable AI approaches
The improvement of wearables allow to collect a high quality data completely automatically, such as the heart rate, activities throughout the day or sleep times. On the other hand, sleep is an important factor for a healthy and happy life and is strongly determined by what is experienced during the day and at night.
In the context of this thesis you should investigate on the basis of self-assessed sleep phases why sleep is better on some days than on others. Moreover, by using state-of-the-art machine learning methods, you should determine whether algorithms can separate good from bad sleep phases. Specifically, you will deep dive into the following steps:
- Use and extend our existing data about sleep behavior and its self-assessment
- Prepare sensor data from wearables such as Garmin smartwatches or Oura Ring for further pre-processing
- Pre-process the available sensor data as features that may have an influence on sleep based on related literature
- Interpretation of the calculated features on the sleep quality with machine learning models and explainable AI approaches
We are interested in understanding how we can most accurately and understandable detect the sleep quality of a person by using wearable devices. Therefore, you should use state-of-the-art machine learning approaches to build robust classifieres. You should start with focussing on conventional approaches (e.g., logistic regression, tree-based models). Furthermore, we expect you to explain why your model outperforms other approaches and thus you should use explainable AI approaches (e.g., statistical approaches or SHAP values) to interpret your classifier.
During the thesis, you will work closely with us and will receive a dedicated supervision. We are highly interested to invest time in your thesis as the topic above counts into our research.
We are interested in understanding how we can most accurately and understandable detect the sleep quality of a person by using wearable devices. Therefore, you should use state-of-the-art machine learning approaches to build robust classifieres. You should start with focussing on conventional approaches (e.g., logistic regression, tree-based models). Furthermore, we expect you to explain why your model outperforms other approaches and thus you should use explainable AI approaches (e.g., statistical approaches or SHAP values) to interpret your classifier.
During the thesis, you will work closely with us and will receive a dedicated supervision. We are highly interested to invest time in your thesis as the topic above counts into our research.
Kevin Koch (kevinkoch@ethz.ch).
Please contact us with your CV, a short statement of motivation, and your current transcripts of records (bachelor & master).
Kevin Koch (kevinkoch@ethz.ch).
Please contact us with your CV, a short statement of motivation, and your current transcripts of records (bachelor & master).