Lorin Werthen-Brabants
FWO Fellow at Ghent University
Postdoctoral researcher at IDLab – Ghent University working on trustworthy machine learning and uncertainty quantification, with applications spanning healthcare diagnostics, time-series analysis, and control.
Selected Publications
- Werthen-Brabants, L., et al. (2024). "Deep Learning-Based Event Counting for Apnea-Hypopnea Index Estimation using Recursive Spiking Neural Networks." IEEE Transactions on Biomedical Engineering.
- Werthen-Brabants, L., et al. (2022). "Split BiRNN for real-time activity recognition using radar and deep learning." Scientific Reports.
- Werthen-Brabants, L., et al. (2022). "Uncertainty quantification for appliance recognition in non-intrusive load monitoring using Bayesian deep learning." Energy and Buildings.
- Bhavanasi, G., Werthen-Brabants, L., et al. (2022). "Patient activity recognition using radar sensors and machine learning." Neural Computing and Applications.
- Castillo-Escario, Y., Werthen-Brabants, L., et al. (2022). "Convolutional neural networks for Apnea detection from smartphone audio signals: effect of window size." IEEE EMBC Conference.
Professional Experience
2025-2028
Postdoctoral Fellow
FWO & IDLab - imec, Ghent University
Focus on Trustworthy Machine Learning (TML) and uncertainty quantification in deep learning models, with applications in healthcare diagnostics. Grant number 1264826N.
2023-2025
Postdoctoral Researcher
IDLab - imec, Ghent University
Research focused on emerging probabilistic hardware (p-bits), with a focus on Model Predictive Control
2019-2023
PhD Student
IDLab - imec, Ghent University
Title of dissertation: "Quantifying Uncertainty and Improving Reliability of Time-Series Based Deep Learning Models".
2018-2019
Machine Learning Engineer
Robovision
Developed machine learning pipelines for computer vision applications, transitioning to academia for deeper exploration of research questions.
Education
2019-2023
PhD in Computer Science
Ghent University
Dissertation: "Quantifying Uncertainty and Improving Reliability of Time-Series Based Deep Learning Models"
Promotors: Prof. dr. ir. Tom Dhaene, Prof. dr. mult. Dirk Deschrijver
2016-2018
MSc in Computer Science
Ghent University
Thesis: Focused on machine learning methods for time-series data.
2013-2016
BSc in Informatics
Ghent University
Grants and Awards
FWO Postdoctoral Fellowship
Funded by the Research Foundation Flanders (FWO) for postdoctoral research on Trustworthy ML and uncertainty quantification in deep learning (2025-2028).
Academic Representation
Selected as a PhD representative for the Flanders AI Research Program (2023).
Invited Talks
Trustworthy and Reliable (Deep) Machine Learning for Healthcare
IBEC, Barcelona, Spain (2025)
Trustworthy ML for Healthcare: Challenges and Developments
Winkelhaak, Antwerp, Belgium (2024)
Teaching and Supervision
Teaching Assistant
Courses on Computer Science, Machine Learning, and Logic at Ghent University (2019-Present).
Master's Thesis Supervision
Supervised multiple Master's theses, with students publishing conference papers under guidance.
Vincent-De Sloover, Louis, et al. "Tailoring Radar-Based Patient Monitoring Models to Real-Life Needs using Utility Maximization." 2022 19th European Radar Conference (EuRAD). IEEE, 2022.
Tuytte, Victor, et al. "Optimized Data Transmission for Radar-Based Edge-Cloud Human Activity Recognition via Quantization." 2024 21st European Radar Conference (EuRAD). IEEE, 2024.
Service and Outreach
Peer Review Service
Regular reviewer for Scientific Reports and IEEE Transactions on Biomedical Engineering.
Science Communication
Featured in public AI discussions, including television appearances (e.g., Karrewiet 2019, VRT NWS Laat 2024).
Mobility and Research Stays
Visiting Researcher
University of Sydney, Australia (2022)
Research Interests
Trustworthy Machine Learning (TML) · Uncertainty Quantification in Deep Learning · Event-Based Time Series Analysis · Multimodal Data Integration · Self-Supervised Learning for Healthcare Applications
Skills
Programming Languages
Deep Learning Frameworks
Data Science
Time Series Analysis, Uncertainty Quantification, Statistical Modeling, Data Visualization.