Evaluation of quality metrics for contact and non-contact physiological signals

Leuven - Master projects/internships
|
More than two weeks ago

During the previous years, imec has been working on the development of technologies for the acquisition of physiological signals from the human body in every-day environments. This includes different form factor of electronic designs that allow to monitor the electrocardiogram (ECG), bioimpedance (BIOZ), among other relevant physiological signals. Since the goal is to allow monitoring with minimal obtrusiveness in everyday life, miniaturized devices and even contactless devices have been developed for this purpose. An important challenge when obtaining signals from real-life environment is the presence of motion artefacts that reduce the usability of the signal or completely distorts it. To face this challenge, imec has been working on quality indicators for ECG and BIOZ, both for signals obtained with contact and contactless devices In order to continue with this work we are looking for students with strong analytic skills, who are creative problem solvers interested in Biomedical signals.

What is the performance of quality estimation methods for contact-based physiological signals (ECG & BIOZ) when applied to non-contact based signals and vice-versa? What features/methods from each domain can be used/merged with features from the other domain? 

 
During the previous years, imec has been working on the development of technologies for the acquisition of physiological signals from the human body in every-day environments. This includes different form factor of electronic designs that allow to monitor the electrocardiogram (ECG), bioimpedance (BIOZ), among other relevant physiological signals. Since the goal is to allow monitoring with minimal obtrusiveness in everyday life, miniaturized devices [i] and even contactless devices [ii,iii] have been developed for this purpose. 

 

An important challenge​ when obtaining signals from real-life environment is the presence of motion artefacts that reduce the usability of the signal or completely distorts it. To face this challenge, imec has been working on quality indicators for ECG and BIOZ, both for signals obtained with contact and contactless [iv] devices. 

 

This internship aims to, after an initial literature review, perform an evaluation of imec’s existing algorithms for quality indication, based on data that has either been already collected or is expected to be collected by the start of the internship. One of the most interesting questions to solve is whether the quality indicators that have been shown to be effective in contact-mode signals are also effective in non-contact mode signals, and vice-versa. 

 

After an initial assessment of the available algorithms for quality estimation, and its performance when applied to the different modalities (contact ECG, non-contact ECG, contact BIOZ, non-contact BIOZ) the student is expected to perform a feature and algorithm optimization, suggesting which methods may be effective in each modality but less effective in other, and which may be applicable to the different modalities (both across signal type and across acquisition technology), suggesting feature/algorithm merging between modalities/technologies when applicable. The internship/thesis work may start by a focus on ECG signals (contact & non-contact) and then move to BIOZ (and possibly other physiological) signals, depending on the time and the findings of the first part of the internship. 


 

 

[1] Imec, accessed: 2018.12.13, http://www.imec-int.com/drupal/sites/default/files/2017-03/HEALTH%20PATCH_1.pdf

[1] Castro, I. D., Morariu, R., Torfs, T., Van Hoof, C., & Puers, R. (2016, May). Robust wireless capacitive ECG system with adaptive signal quality and motion artifact reduction. In Medical Measurements and Applications (MeMeA), 2016 IEEE International Symposium on (pp. 1-6). IEEE.

[1] Torfs, T., Chen, Y. H., Kim, H., & Yazicioglu, R. F. (2014). Noncontact ECG recording system with real time capacitance measurement for motion artifact reduction. IEEE transactions on biomedical circuits and systems, 8(5), 617-625.

[1] Castro, I. D., Varon, C., Torfs, T., Van Huffel, S., Puers, R., & Van Hoof, C. (2018). Evaluation of a Multichannel Non-Contact ECG System and Signal Quality Algorithms for Sleep Apnea Detection and Monitoring. Sensors, 18(2), 577.​


 


Type of project: Internship, Combination of internship and thesis

Duration: 6-9 months

Required degree: Master of Science, Master of Engineering Science

Required background: Biomedical engineering, Computer Science, Electrotechnics/Electrical Engineering

Supervising scientist(s): For further information or for application, please contact: Ivan Dario Castro Miller (ivand.castro@imec.be) and Neide Simoes Capela (Neide.Simoes@imec.be)

Allowance only for students from a non-Belgian university

Share this on

truetrue

This website uses cookies for analytics purposes only without any commercial intent. Find out more here. Our privacy statement can be found here. Some content (videos, iframes, forms,...) on this website will only appear when you have accepted the cookies.

Accept cookies