Jun 2025–present
Assistant Professor
Delft University of Technology
Department of Engineering Systems and Services, Faculty of Technology, Policy and Management.
Curriculum vitae
Benjamin Filtjens
Assistant Professor · Delft University of Technology
Academic CV
Research profile
I am an Assistant Professor at TU Delft developing machine learning methods to understand and improve human movement across clinical sensing, rehabilitation and assistive robotics. During my PhD at KU Leuven, I initiated a collaboration between machine learning and movement-disorder neuroscience to develop objective methods for assessing pathological gait. A KU Leuven postdoctoral fellowship expanded this work towards multimodal sensing and adaptive assistive robotics.
A subsequent Data Sciences Institute fellowship at the University of Toronto supported my project, “Towards foundation models for gait analysis”. Working with KITE and the Vector Institute shifted my focus towards movement representations that generalise across datasets, populations and tasks. In 2025, I received the International Society of Posture and Gait Research Emerging Scientist Award.
I have built an interdisciplinary research programme connecting machine learning, neuroscience, physiotherapy and rehabilitation robotics. I co-wrote AID-FOG, RevalExo, AI@WZC, AidWear and JustHealth and am a co-investigator, generally leading or co-leading their machine learning research. Across these collaborations, I am the daily scientific advisor to five PhD candidates.
My ambition is to establish new principles for foundation learning from heterogeneous, partially connected data and translate them into AI for clinical assessment and adaptive rehabilitation, with the goal of improving patients’ quality of life.
Academic appointments
Oct 2024–present
Research Fellow, voluntary appointment
KU Leuven, Department of Electrical Engineering (ESAT)
10% unpaid scientific affiliation to coordinate ongoing collaborations.
Oct 2024–Jun 2025
Postdoctoral Fellow
KITE Research Institute, University Health Network, and Vector Institute, Canada
Data Sciences Institute fellowship, University of Toronto. Mentor: Prof. Babak Taati.
Oct 2023–Sep 2024
Postdoctoral Researcher
KU Leuven, Department of Electrical Engineering (ESAT)
Research with Prof. Bart Vanrumste.
Oct 2022–Sep 2023
Postdoctoral Fellow
KU Leuven, Department of Electrical Engineering (ESAT)
KU Leuven Research Council Postdoctoral Mandate. Mentor: Prof. Bart Vanrumste.
Sep 2017–Sep 2022
Doctoral Researcher
KU Leuven, Department of Electrical Engineering (ESAT)
Deep learning for automated analysis of pathological gait.
Education
2022
PhD in Electrical Engineering
KU Leuven
Awarded 9 September 2022. Thesis: Deep learning for automated movement analysis using motion capture: case studies on detecting and assessing gait impairments in people with Parkinson’s disease and freezing of gait. Promotor: Prof. Bart Vanrumste; co-promotors: Prof. Alice Nieuwboer and Prof. Peter Slaets.
2017
MSc in Engineering Technology, Automation
Hasselt University and KU Leuven
Double degree. Thesis: Detection of Lint by Using Machine Vision.
Personal grants and fellowships
2024–2025
Data Sciences Institute Postdoctoral Fellowship
University of Toronto
Towards foundation models for gait analysis. Fellowship DSI-PDFY3R1P13.
2022–2023
Research Council Postdoctoral Mandate
KU Leuven
Towards at-home mobility assessment with deep learning technology. Fellowship PDMT2/22/046.
Awards and recognition
2025
Emerging Scientist Award
International Society of Posture and Gait Research
Recognised for early-career contributions to posture and gait research; delivered the Emerging Scientist plenary lecture at the 2025 World Congress.
Collaborative research projects
2024
AID-FOG
Michael J. Fox Foundation for Parkinson’s Research
Co-investigator; machine learning research. Artificial intelligence-driven freezing of gait detection in the home.
2023
RevalExo
Research Foundation – Flanders (FWO), Strategic Basic Research
Co-investigator; machine learning research. A wearable lower-limb exoskeleton for daily activities and data-driven remote rehabilitation.
2023
AI@WZC
Province of Flemish Brabant, Smart Region
Project conception, proposal writing, and scientific coordination. Supporting and motivating care-home residents to exercise using artificial intelligence.
2023
AidWear
Belgian Federal Public Service Policy and Support (BOSA)
Co-investigator; machine learning research. AI-driven wearable robotics for healthcare.
2026
JustHealth
European Union, Horizon Europe, Marie Skłodowska-Curie Actions
Co-investigator and daily scientific advisor. A doctoral network on just and trustworthy artificial intelligence in healthcare.
2021
FOG-IT
KU Leuven Industrial Research Fund
Research contributor. An integrated software platform for tagging freezing of gait in Parkinson’s disease.
Doctoral supervision
Current
Daily scientific advisor
KU Leuven collaborations
Juha Carlon, Diwas Lamsal, Vayalet Stefanova, Maxim Yudayev, and Y. Haghbin (started September 2026). Research spans pathological gait, multimodal sensing, rehabilitation, and assistive technology.
Academic service
2025–present
Academic Editor
PLOS Digital Health
Editorial service in digital health research.
2026
Program Committee member
ECCV MoCha Workshop
Human Motion Challenges in Real-World and Clinical Settings.
Inaugural member, Measurement Group
International Consortium for Freezing of Gait (ICFOG)
Contributed to international recommendations on technology-assisted freezing-of-gait measurement and machine learning evaluation.
Scientific advisory committee
N-AI-tingale, FWO-funded project
Expertise in audio-visual fusion for identifying clinical decline in hospitalised patients.
Ambassador
TU Delft Institute for Health Systems Science
Connecting methodological and clinical research.
Technical Committee member
IEEE EMBS Benelux, BioRobotics
Interdisciplinary community service in rehabilitation and assistive robotics.
Open research resources
2025
CARE-PD
Multisite clinical gait dataset
Contribution to data sharing, harmonisation, pretraining, and release.
2026
RevalExo
Multimodal locomotion benchmark and dataset
Co-design of a temporal, cross-population benchmark for locomotion mode recognition.
2026
HERMES
Open-source sensing and AI framework
Co-design and scientific supervision of synchronised multimodal sensing, real-time inference, and intervention.
Patent applications
2026
System for assessment of freezing of gait
Pending patent application PCT/EP2026/065548
Co-inventor with P.-K. Yang, B. Vanrumste, M. Gilat, and P. Ginis. Inventor order is not final.
Selected publications
Five selected publications. See the full publication list for all publications and research outputs.
2026
Conference Paper
CARE-PD: A multi-site anonymized clinical dataset for parkinson’s disease gait assessment
Advances in Neural Information Processing Systems 38, 2026
Citation
Adeli V, Klabučar I, Rajabi J, Filtjens B, Mehraban S, Wang D, Hoang TH, Do M, Seo H, Muller C, Coelho D, de Oliveira C, Ginis P, Gilat M, Nieuwboer A, Spildooren J, Mckay J, Kwon H, Clifford G, Esper C, Factor S, Genias I, Dadashzadeh A, Shum L, Whone A, Mirmehdi M, Iaboni A, Taati B (2026). CARE-PD: A multi-site anonymized clinical dataset for parkinson’s disease gait assessment. Advances in Neural Information Processing Systems 38, 2026. https://doi.org/10.52202/085713-4391
2026
Journal Article
Deep learning for freezing of gait assessment using inertial measurement units: a multicentre validation study
npj Parkinson's Disease, 2026
Citation
Yang PK, Carlon J, Goris M, Klaver E, Nonnekes J, van Wezel RJA, Alcock L, Yarnall AJ, Rochester L, Hansen C, Schlenstedt C, Maetzler W, Buzaglo D, Brozgol M, Hausdorff JM, Nieuwboer A, Gilat M, Ginis P, Vanrumste B, Filtjens B (2026). Deep learning for freezing of gait assessment using inertial measurement units: a multicentre validation study. npj Parkinson's Disease, 2026. https://doi.org/10.1038/s41531-026-01407-7
2026
Preprint
RevalExo: A Functional Daily-Activity Benchmark for Inertial and Visual Locomotion Mode Recognition in Older Adults and Clinical Cohorts
arXiv preprint arXiv:2609.08090, 2026
BMVC 2026, oral presentation, in press; linked version is a preprint.
Citation
Lamsal D, Carlon J, Claeys R, Yudayev M, Flynn L, Verstraten T, Beckwée D, Swinnen E, Bâce M, Vanrumste B, Filtjens B (2026). RevalExo: A Functional Daily-Activity Benchmark for Inertial and Visual Locomotion Mode Recognition in Older Adults and Clinical Cohorts. arXiv preprint arXiv:2609.08090, 2026. https://doi.org/10.48550/ARXIV.2609.08090
2026
Journal Article
Assessment of change in drug-induced parkinsonism in daily-life gait using vision-based machine learning
Corresponding authors: Filtjens B and Iaboni A.
PLOS Digital Health
In press.
Citation
Filtjens B, Adeli V, Mehraban S, Sabo A, Taati B, Iaboni A (2026). Assessment of change in drug-induced parkinsonism in daily-life gait using vision-based machine learning. PLOS Digital Health. https://doi.org/10.1371/journal.pdig.0001780 Corresponding authors: Filtjens B and Iaboni A.
2022
Journal Article
Skeleton-based action segmentation with multi-stage spatial-temporal graph convolutional neural networks
IEEE Transactions on Emerging Topics in Computing 12 (1), 202-212, 2022
Citation
Filtjens B, Vanrumste B, Slaets P (2022). Skeleton-based action segmentation with multi-stage spatial-temporal graph convolutional neural networks. IEEE Transactions on Emerging Topics in Computing 12 (1), 202-212, 2022. https://doi.org/10.1109/TETC.2022.3230912