Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Sleep Diagnoses Classification Based On Longitudinal Wearable Sensors Data
This project aims to extract features from wearable sensor data. Longitudinal recordings for up to one year will be provided for hundreds of patients with sleep disorders.
From 4% to 6% of the general population suffers from hypersomnolence. This is a sleep disorder characterized by an excessive need for sleep day and night. People with hypersomnolence may be diagnosed with narcolepsy type 1, characterized by cataplexy (loss of muscle tone). In contrast, precise diagnostic biomarkers are lacking for other patients using standard clinical assessments such as polysomnography in the sleep laboratory and questionnaires. Making a correct diagnosis is a problem for patients and physicians, resulting in difficulty in treating the disease. Thanks to advances in wearable technologies, it is now possible to monitor sleep, physical activity, and physiological parameters for many consecutive days at patients' homes. Therefore, the present study aims to identify new digital biomarkers by analyzing longitudinal data from wearable sensors in narcolepsy and its borderland using a smartwatch device for one year.
From 4% to 6% of the general population suffers from hypersomnolence. This is a sleep disorder characterized by an excessive need for sleep day and night. People with hypersomnolence may be diagnosed with narcolepsy type 1, characterized by cataplexy (loss of muscle tone). In contrast, precise diagnostic biomarkers are lacking for other patients using standard clinical assessments such as polysomnography in the sleep laboratory and questionnaires. Making a correct diagnosis is a problem for patients and physicians, resulting in difficulty in treating the disease. Thanks to advances in wearable technologies, it is now possible to monitor sleep, physical activity, and physiological parameters for many consecutive days at patients' homes. Therefore, the present study aims to identify new digital biomarkers by analyzing longitudinal data from wearable sensors in narcolepsy and its borderland using a smartwatch device for one year.
This project aims to extract relevant features from longitudinal wearable sensor data. The student will be able to work with data already collected in a real clinical and ambulatory setting with patients suffering from hypersomnolence due to narcolepsy, psychiatric disorders, or substance intake. Diagnoses ground truth data will be provided from the clinical database.
WORK PACKAGES:
- WP1: Literature research on potential biomarkers for hypersomnolence and disease classification algorithms.
- WP2: Time series signal processing and features extraction.
- WP3: Algorithm for diagnoses classification.
This project aims to extract relevant features from longitudinal wearable sensor data. The student will be able to work with data already collected in a real clinical and ambulatory setting with patients suffering from hypersomnolence due to narcolepsy, psychiatric disorders, or substance intake. Diagnoses ground truth data will be provided from the clinical database.
WORK PACKAGES: - WP1: Literature research on potential biomarkers for hypersomnolence and disease classification algorithms. - WP2: Time series signal processing and features extraction. - WP3: Algorithm for diagnoses classification.
Work with a real dataset collected in an unsupervised setting, the home of patients with diagnosed sleep and behavioral disorders.
Contribute to developing an algorithm to find digital biomarkers that could help the patients and the physician better understand the disease and the proper treatment.
The project is led and supervised by experts in data analysis, machine learning, and experts from the clinical field. You can benefit from mentoring in both areas.
Work with a real dataset collected in an unsupervised setting, the home of patients with diagnosed sleep and behavioral disorders.
Contribute to developing an algorithm to find digital biomarkers that could help the patients and the physician better understand the disease and the proper treatment.
The project is led and supervised by experts in data analysis, machine learning, and experts from the clinical field. You can benefit from mentoring in both areas.
- Strong programming skills in Python.
- Strong background in machine learning.
- Structured and reliable working style.
- Ability to work independently on a challenging topic.
- Proven records on some of the following: experience in large data sets, longitudinal data analysis, and sparse feature selection.
- Strong programming skills in Python. - Strong background in machine learning. - Structured and reliable working style. - Ability to work independently on a challenging topic. - Proven records on some of the following: experience in large data sets, longitudinal data analysis, and sparse feature selection.
If interested, please contact me via email with your CV and transcripts attached: Oriella Gnarra oriella.gnarra@hest.ethz.ch
If interested, please contact me via email with your CV and transcripts attached: Oriella Gnarra oriella.gnarra@hest.ethz.ch