Stéphanie DEVUYST PhD - Electrical Engineer - Teaching Assistant stephanie.devuyst@umons.ac.be Faculty of Engineering, Mons Circuit Theory and Signal Processing Division

Profile

---

I was born in Namur, Belgium, in 1980. In 2003, I hold an electrical engineering degree from the Faculté Polytechnique de Mons (FPMS, Belgium). During 2 years, I worked on the WALEO project DREAMS. This project was dedicated to automatically determine the sleep stages which follow during one night by analyzing polysomnographics signals (EOG, EEG, EMG, etc.). In this way, I started a PhD thesis on this subject. Since 2011, I hold my PhD in applied sciences from the University of Mons on biomedical signal processing. Now, I'm an assistant in the TCTS Lab of the FPMs. I take part to the Circuit Theory and Signal Processing Labs and exercises. I also supervise several student projects in second and final year.

Research Interests

---

My current research activities are focusing on automatic sleep analysis and biomedical signal processing. I was involved in the WALEO project DREAMS which aimed to analyze polysomnographics signals (EOG, EEG, EMG, etc.) to determine the sleep stages which follow during one night. More especially, we took better account of the artifacts present in the signals in order to improve the sleep stage quotation.

I'm also a member of the Biomedical Group of the TCTS Lab.

Academic Links

---

• FPMs: Faculté Polytechnique de Mons - Belgium
• TCTS Lab: Circuit Theory and Signal Processing Lab
• Biomed: Biomedical Data Processing Group
• Dreams: research project for automatic sleep stage scoring
• TI: The pole of excellence "Technologies de l'information" of the Faculté Polytechnique de Mons
• IAP DYSCO: Interuniversity Attraction Poles on Dynamical systems, control and optimization

Publications

---

PhD Dissertation

2011

• S. Devuyst., "Analyse Automatique de Tracés Polysomnographiques d’Adultes", Thèse de Doctorat en Sciences de l’Ingénieur (PhD. Thesis in Applied Sciences), Promoteur: Prof. T. Dutoit, Faculté Polytechnique de Mons de l’université de Mons (UMONS), Belgique, 2011.

Regular Papers in Journals

2008

• S. Devuyst, T. Dutoit, P. Stenuit, M. Kerkhofs, E. Stanus, "Canceling ECG Artifacts in EEG using a Modified Independent Component Analysis Approach", EURASIP Journal on Advances in Signal Processing, Volume 2008, Article ID 747325, Accepted 31 July 2008.

Papers in Conference Proceedings

2011

• S. Devuyst, T. Dutoit, P. Stenuit, M. Kerkhofs, "Automatic Sleep Spindles Detection - Overview and Development of a Standard Assessment Method", Proceedings of the 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC 2011), WeC19.4, pp 1713-1716, Boston, Aug 30-Sep 3, 2011.
• J. Taelman, B. Mijovic, S. Devuyst, T. Dutoit, S. Van Huffel, "ECG artifact removal from surface EMG signals by combining empirical mode decomposition and independent component analysis", Proceedings of the International Conference on Bio-inspired Systems and Signal Processing (BIOSIGNALS2011), subconference of the 4th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2011), pp 421-424, Rome, Italy, 26-29 january, 2011.

2010

• S. Devuyst, T. Dutoit, P. Stenuit, M. Kerkhofs, "Automatic K-complexes Detection in Sleep EEG Recordings using Likelihood Thresholds", Proceedings of the 32nd Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC 2010), FrDPo04.6, pp 4658-4661, Buenos Aires, Argentina, Aug 31-Sept 4, 2010.
• S. Devuyst, T. Dutoit, T. Ravet, P. Stenuit, M. Kerkhofs, "Comparison of Visual Sleep Stage Classification according to AASM and Rechtschaffen & Kales Rules", Journal of Sleep Research, Vol.19, Suppl.2, pp 358, Abstract P1039 of the 20th Congress of the European Sleep Research Society (ESRS 2010), Lisbone Portugal, 14-18 Sept, 2010.

2008

• S. Devuyst, T. Dutoit, T Ravet, P. Stenuit, M. Kerkhofs, E. Stanus, "Automatic Processing of EEG-EOG-EMG Artifacts in Sleep Stage Classification", Proceedings 23 of the 13th International Conference on BioMedical Engineering (ICBME08), Session 2B–3: Biosignal Processing IV, pp 146–150, Singapore, 3rd - 6th December, 2008.
• S. Devuyst, T. Dutoit, P. Stenuit, M. Kerkhofs, E. Stanus, "Removal of ECG Artifacts from EEG using a Modified Independent Component Analysis Approach", Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC 2008), SaET1.1, pp 5204-5207, Vancouver Aug 20-24, 2008.

2006

• S. Devuyst, T. Dutoit, JF. Didier, F. Meers, E. Stanus, P. Stenuit, M. Kerkhofs, "Automatic Sleep Spindle Detection in Patients with Sleep Disorders", Proceedings of the 28th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC 2006), FrEP5.5, New York Aug 30-Sept 3, 2006.
• J. Tilmanne, P. Stenuit, M. Kerkhofs, S. Devuyst, T. Dutoit, "Automatic scoring of periodic limb movements using a threshold based algorithm", 18th Congress of the European Sleep Research Society, Innsbruck, Austria. 2006.
• J. Tilmanne, P. Stenuit, M. Kerkhofs, S. Devuyst, T. Dutoit, "Automatic scoring of periodic limb movements using a threshold based algorithm.", Autumn meeting of the Belgian Association for Sleep research and Sleep medicine (BASS), Gent, Belgium , 2006.

2004

• S. Devuyst, T.Dutoit, B. Arslan, O.Bilenne, F. Meers, E.Stanus, P.Stenuit, M.Kerkhofs, "Artifacts processing for sleep stage classification", URSI forum broadband communications, pp 53, 10 decembre 2004.

Thesis

2011

• S. Devuyst., "Analyse Automatique de Tracés Polysomnographiques d’Adultes", Thèse de Doctorat en Sciences de l’Ingénieur (PhD. Thesis in Applied Sciences), Promoteur: Prof. T. Dutoit, Faculté Polytechnique de Mons de l’université de Mons (UMONS), Belgique, 2011.

2007

• S. Devuyst., "Classification automatique en stades du sommeil : extraction de caractéristiques et comparaison des principaux classificateurs", Rapport de DEA (Master's thesis in applied sciences), Faculté Polytechnique de Mons, 2007.

2003

• S. Devuyst., "Reconnaissance automatique de partitions manuscrites", Travail de fin d'étude (Master's thesis in applied sciences), Faculté Polytechnique de Mons, 2003.

Other

2004

• S. Devuyst., "Une analyse automatique du sommeil? Vous ne rêvez pas...", Polytech-News, journal interne de la Faculté Polytechnique de Mons, 12ème année, No 26, Février - mars - avril 2004.

The DREAMS Databases

---

During the DREAMS project, we collected a large amount of polysomnographic recordings (PSG) to tune, train and test our automatic detection algorithms.

These recordings were annotated in microevents or in sleep stages by several experts. They were acquired in a sleep laboratory of a belgium hospital using a digital 32-channel polygraph (BrainnetTM System of MEDATEC, Brussels, Belgium). The standard European Data Format (EDF) was used for storing.

In order to facilitate future research and performance comparision, we decided to publish these data on Internet. Therefore, eight DREAMS databases are available according to the annotation carried out (click on the link to open):

• The DREAMS Subjects Database: 20 whole-night PSG recordings coming from healthy subjects, annoted in sleep stages according to both the Rechtschaffen and Kales criteria and the new standard of the American Academy of Sleep Medicine;


• The DREAMS Patients Database: 27 whole-night PSG recordings coming from patients with various pathologies, annoted in sleep stages according to both the Rechtschaffen and Kales criteria and the new standard of the American Academy of Sleep Medicine;


• The DREAMS Artifacts Database: 20 excerpts of 15 minutes of PSG recordings annoted in artifacts (cardiac interference, slow ondulations, muscle artifacts, failing electrode, 50/60Hz main interference, saturations, abrupt transitions, EOG interferences and artifacts in EOG) by an expert;


• The DREAMS Sleep Spindles Database: 8 excerpts of 30 minutes of central EEG channel (extracted from whole-night PSG recordings), annotated independently by two experts in sleep spindles;


• The DREAMS K-complexes Database: 5 excerpts of 30 minutes of central EEG channel (extracted from whole-night PSG recordings), annotated independently by two experts in K-complexes;


• The DREAMS REMs Database: 9 excerpts of 30 minutes of PSG recordings in which rapid eye movements were annotated by an expert;


• The DREAMS PLMs Database: 10 whole-night PSG recordings coming from patients in which one of the two tibialis EMG was annoted in periodic limb movements by an expert;


• The DREAMS Apnea Database: 12 whole-night PSG recordings coming from patients annoted in respiratory events (central, obstructive and mixed apnea and hypopnea) by an expert.

These Dreams Databases of University of MONS - TCTS Laboratory (Stéphanie Devuyst, Thierry Dutoit) and Université Libre de Bruxelles - CHU de Charleroi Sleep Laboratory (Myriam Kerkhofs) is under terms of the Attribution-NonCommercial-NoDerivs 3.0 Unported (CC BY-NC-ND 3.0) License. Each use of one of these databases and/or its content must be attributed to University of MONS - TCTS Laboratory (Stéphanie Devuyst, Thierry Dutoit) and Université Libre de Bruxelles - CHU de Charleroi Sleep Laboratory (Myriam Kerkhofs). Any other additional authorisations may be asked to avre@umons.ac.be.

Assessment Algorithm

---

During the DREAMS project, we also observed that perfomances completed in the same field by other authors were hardly comparable because their assessment methods were radically different. That why we propose a unique assessment method using a well defined terminology and from which it is possible to establish all the desired confusion matrices. This assessment method allows to compare an automatic annotation in micro-events with one or two visual annotations. It has been implemented under Matlab and the corresponding algorithms can be downloaded Here.

Contact

---