Gem job

DTU Studieprojekt - MSc thesis - Mental states classification from unlabelled EEG data using self-supervised learning tools

MSc thesis - Mental states classification from unlabelled EEG data using self-supervised learning tools

Udbyder
Vejleder
Sted
København og omegn
The objective of this project is to implement a self-supervised method to classify mental workload in an everyday setting using our Nexus device.The project will begin with a review of self-supervised learning, specifically applied to EEG signals and other biological signals. The next step will be to reproduce the results obtained by Bainville et al. using publicly available datasets [1]. Next, a small experimental protocol will be designed to generate data using the Nexus device. This process will involve determining a task (i.e. Meditation, Mental workload, Eyes closed/open, etc.), the minimum number of samples/recordings, the pretext tasks (Relative positioning, temporal shuffling, etc.) and other optimal parameters for improving the self-supervised learning capabilities. Lastly, the performance will be compared to other baseline methods and will be thoroughly analyzed to determine which factors affect the performance.

Project recommended requirements:

? Experience with EEG signal processing and analysis.

? Experience with Python

? Experience with ML/DL

? Experience with Self-supervised learning

References

[1] Banville, Hubert, et al. "Uncovering the structure of clinical EEG signals with self-supervised learning." Journal of Neural Engineering 18.4 (2021): 046020.

I samarbejde med
Mark Yousef (Insai aps)

Forudsætninger
programming

Emneord

Tags
Kontakt
Virksomhed/organisation
DTU Elektro

Navn
Silvia Tolu

Stilling
Lektor

Mail
stolu@elektro.dtu.dk

Vejleder-info
Kandidatuddannelsen i Menneskeorienteret Kunstig Intelligens
Vejleder
Silvia Tolu

Kandidatuddannelsen i Bioinformatik og Systembiologi
Vejleder
Silvia Tolu

Kandidatuddannelsen i Medicin og Teknologi
Vejleder
Silvia Tolu

Kandidatuddannelsen i Informationsteknologi
Vejleder
Silvia Tolu

Kandidatuddannelsen i Elektroteknologi
Vejleder
Silvia Tolu

Kandidatuddannelsen i Fotonik
Vejleder
Silvia Tolu

Kandidatuddannelsen i Matematisk Modellering og Computing
Vejleder
Silvia Tolu

Kandidatuddannelsen i Autonome Systemer
Vejleder
Silvia Tolu

Kandidatuddannelsen i Fysik og Nanoteknologi
Vejleder
Silvia Tolu

Skriv i din ansøgning, at du fandt jobbet på ofir.dk


ellerGem job

Du har ikke vedlagt dokumenter til din ansøgning. Klik på 'OK' knappen nedenfor for at sende din ansøgning alligevel, eller klik på knappen "Annuller" og vedhæft dokumenter, før du sender din ansøgning igen.

Modtag jobannoncer der minder om dette job i din indbakke.Når du tilmelder dig accepterer du samtidig vores privatlivspolitik. Du kan altid trække dit samtykke tilbage.

DTU Studieprojekt - MSc thesis - Mental states classification from unlabelled EEG data using self-supervised learning tools

MSc thesis - Mental states classification from unlabelled EEG data using self-supervised learning tools

Udbyder
Vejleder
Sted
København og omegn
The objective of this project is to implement a self-supervised method to classify mental workload in an everyday setting using our Nexus device.The project will begin with a review of self-supervised learning, specifically applied to EEG signals and other biological signals. The next step will be to reproduce the results obtained by Bainville et al. using publicly available datasets [1]. Next, a small experimental protocol will be designed to generate data using the Nexus device. This process will involve determining a task (i.e. Meditation, Mental workload, Eyes closed/open, etc.), the minimum number of samples/recordings, the pretext tasks (Relative positioning, temporal shuffling, etc.) and other optimal parameters for improving the self-supervised learning capabilities. Lastly, the performance will be compared to other baseline methods and will be thoroughly analyzed to determine which factors affect the performance.

Project recommended requirements:

? Experience with EEG signal processing and analysis.

? Experience with Python

? Experience with ML/DL

? Experience with Self-supervised learning

References

[1] Banville, Hubert, et al. "Uncovering the structure of clinical EEG signals with self-supervised learning." Journal of Neural Engineering 18.4 (2021): 046020.

I samarbejde med
Mark Yousef (Insai aps)

Forudsætninger
programming

Emneord

Tags
Kontakt
Virksomhed/organisation
DTU Elektro

Navn
Silvia Tolu

Stilling
Lektor

Mail
stolu@elektro.dtu.dk

Vejleder-info
Kandidatuddannelsen i Menneskeorienteret Kunstig Intelligens
Vejleder
Silvia Tolu

Kandidatuddannelsen i Bioinformatik og Systembiologi
Vejleder
Silvia Tolu

Kandidatuddannelsen i Medicin og Teknologi
Vejleder
Silvia Tolu

Kandidatuddannelsen i Informationsteknologi
Vejleder
Silvia Tolu

Kandidatuddannelsen i Elektroteknologi
Vejleder
Silvia Tolu

Kandidatuddannelsen i Fotonik
Vejleder
Silvia Tolu

Kandidatuddannelsen i Matematisk Modellering og Computing
Vejleder
Silvia Tolu

Kandidatuddannelsen i Autonome Systemer
Vejleder
Silvia Tolu

Kandidatuddannelsen i Fysik og Nanoteknologi
Vejleder
Silvia Tolu

Skriv i din ansøgning, at du fandt jobbet på ofir.dk


+
Gem

Mere af samme slags?

Modtag jobannoncer der minder om dette job i din indbakke.




Når du tilmelder dig accepterer du samtidig vores privatlivspolitik. Du kan altid trække dit samtykke tilbage.