Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Machine Learning for Healthcare: Signal Quality Assessment of Biosignals
We aim to develop an efficient algorithm able to run in real-time within a mobile phone, to collect high-quality electrocardiogram (ECG) and photoplethysmogram (PPG) signals.
Keywords:
The purpose of this study is to develop an algorithm that automatically indicates if the collected electrocardiogram (ECG) and photoplethysmogram (PPG) signals are of adequate quality for clinical interpretation and diagnosis. The algorithm also needs to be able to identify common causes of low signal-to-noise ratio in ECG and PPG signals, such as misplaced electrodes, poor skin conductivity, external electrical interference, and artifacts resulting from patient motion.
The purpose of this study is to develop an algorithm that automatically indicates if the collected electrocardiogram (ECG) and photoplethysmogram (PPG) signals are of adequate quality for clinical interpretation and diagnosis. The algorithm also needs to be able to identify common causes of low signal-to-noise ratio in ECG and PPG signals, such as misplaced electrodes, poor skin conductivity, external electrical interference, and artifacts resulting from patient motion.
- Motion artifact identification
- Lead misplacement identification
- User interface development
**Tasks**
- Literature review (10%)
- Data analysis (loading data) (10%)
- Design and implement an optimal filter for each biosignal noise type (30%)
- Design and implement a multiclass machine learning classifier (30%)
- Test, compare and evaluate results from different noise types (10%)
- Report and present results (10%)
- Motion artifact identification - Lead misplacement identification - User interface development
**Tasks**
- Literature review (10%) - Data analysis (loading data) (10%) - Design and implement an optimal filter for each biosignal noise type (30%) - Design and implement a multiclass machine learning classifier (30%) - Test, compare and evaluate results from different noise types (10%) - Report and present results (10%)
Dr Moe Elgendi (moe.elgendi@hest.ethz.ch) will supervise the student at the Biomedical and Mobile Health Technology Research Group in ETH Zurich’s D-HEST Department of Health Sciences and Technology.
Google Scholar: https://scholar.google.com/citations?user=-WFwzjoAAAAJ&hl=en
Researchgate: https://www.researchgate.net/profile/Mohamed-Elgendi
Dr Moe Elgendi (moe.elgendi@hest.ethz.ch) will supervise the student at the Biomedical and Mobile Health Technology Research Group in ETH Zurich’s D-HEST Department of Health Sciences and Technology.
Google Scholar: https://scholar.google.com/citations?user=-WFwzjoAAAAJ&hl=en