Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Extracting respiratory rate from photoplethysmogram using machine learning algorithms
How reliably and accurately can a person's respiratory rate be estimated from wearable pulse recordings?
Keywords: Physiological sensing, Machine learning
Respiratory rate (RR) is one of the primary vital signs and crucial to assess a person's state of health, but current assessment methods require covering the mouth and nose with a mask.
In this project, we will extract the RR from a person's photoplethysmogram (PPG) - a vital signal that is routinely and conveniently recorded on today's smartwatches and fitness bands. We will develop a machine learning-based approach to extract the RR from PPG signals that achieves the reliability and accuracy needed for clinical requirements. We will work with available datasets and focus on state-of-the-art deep learning algorithms. To ensure the reliability of our results, we will investigate probabilistic neural networks to predict the confidence intervals of the predicted RRs.
_Note: Candidates should be experienced in machine learning and corresponding toolchains (e.g., PyTorch or Tensorflow), ideally being familiar with methods in Computer Vision. Experience in signal processing is beneficial but not required._
Respiratory rate (RR) is one of the primary vital signs and crucial to assess a person's state of health, but current assessment methods require covering the mouth and nose with a mask.
In this project, we will extract the RR from a person's photoplethysmogram (PPG) - a vital signal that is routinely and conveniently recorded on today's smartwatches and fitness bands. We will develop a machine learning-based approach to extract the RR from PPG signals that achieves the reliability and accuracy needed for clinical requirements. We will work with available datasets and focus on state-of-the-art deep learning algorithms. To ensure the reliability of our results, we will investigate probabilistic neural networks to predict the confidence intervals of the predicted RRs.
_Note: Candidates should be experienced in machine learning and corresponding toolchains (e.g., PyTorch or Tensorflow), ideally being familiar with methods in Computer Vision. Experience in signal processing is beneficial but not required._
Not specified
- Björn Braun, bjoern.braun@inf.ethz.ch
- Manuel Meier, manuel.meier@inf.ethz.ch
- Björn Braun, bjoern.braun@inf.ethz.ch - Manuel Meier, manuel.meier@inf.ethz.ch