Turning sensor data into
evidence you can act on.
Postdoctoral researcher, Maastricht University (NUTRIM) PhD in Biomedical Engineering, Technion Licensed physiotherapist (Israel)
I'm looking for Application Science, Clinical & Scientific Affairs, Evidence Generation and R&D roles in wearable sensors, digital health and assistive technology. My background spans clinical practice, study design and reproducible analysis tooling across IMU, EMG, PPG and accelerometry, from single patients to 10,000-participant cohorts.
Multi-Sensor Measurement Systems
IMU, EMG, PPG/ECG, and accelerometry — from single-device setups to synchronized multi-sensor protocols. Practical experience with placement, synchronization, artifacts, drift, and emerging form factors including earables and smart glasses.
Evidence Generation & Study Design
End-to-end study execution: protocols, endpoints, feasibility, adherence, and data-quality checks. Experience across observational studies, RCTs, and longitudinal designs, translating questions into testable, defensible outcomes.
Human Movement & Clinical Domain
Five years of clinical physiotherapy plus a PhD in biomedical engineering. Deep knowledge of gait, functional movement, neuromuscular physiology, and what sensor signals can (and cannot) tell you about real humans.
Real-World Data & Population Analytics
Working with large wearable datasets (10,000+ participants) to derive interpretable activity-pattern metrics — bouts, fragmentation, transitions, intensity distribution — and relate them to health outcomes.
Validation & Regulatory-Ready Measurement
Gold-standard comparisons, reliability testing, synchronization, artifact handling, and transparent error reporting. Building the evidence base so performance claims are defensible for product and regulatory decisions.
Clinical-Technical Translation
Bridging clinical needs and technical implementation. Previously embedded in a product team as clinical consultant, translating end-user requirements into engineering specifications. Comfortable in both languages.
Selected work
Spanning population health, device validation, multi-sensor fusion, and clinical measurement.
Population-scale physical activity analytics
Challenge: Large accelerometry datasets are messy, and summary metrics like "steps/day" miss meaningful behavior patterns.
What I did: Built reproducible Python workflows to quantify activity-pattern metrics (bout distributions, fragmentation, transitions, intensity distribution) from The Maastricht Study's activPAL dataset (~10,000 participants).
Output: Analysis-ready measures linked to cardiometabolic outcomes.
REALM: Real-world ecological movement with synchronized sensors
Challenge: Current wearable approaches capture the "what" of movement but not the context — you can detect slow walking, but not whether it's caused by cognitive load, stairs ahead, or another pedestrian.
What I did: Co-designed a funded multi-sensor protocol (thigh-worn IMU + ear-worn IMU + smart glasses with egocentric video) to capture movement in real-world conditions: clutter, stairs, turns, social navigation, and dual-task cognitive load.
Output: Public dataset and open-source analysis pipeline for sensor fusion benchmarking. Currently in data collection.
End-to-end wearable gait assessment after ACL reconstruction
Challenge: Lab snapshots miss day-to-day movement strategies during rehabilitation. Wearable-derived endpoints need the same rigor as lab measures.
What I did: Designed a wearable IMU workflow for multi-pace gait and stair tasks in real-world settings.
Ran a randomized sham-controlled trial of vibrational stimulation post-surgery.
Published the full dataset (COMPWALK-ACL (opens in new tab) ) in Scientific Data (opens in new tab) for open reuse.
Output: 5 peer-reviewed papers, 1 open dataset, reusable pipeline components. Protocol to publication, end-to-end.
Portable device validation for clinical measurement
Challenge: Can accessible, portable devices produce research-grade measurements with standardized protocols?
What I did: Designed and ran inter- and intra-session reliability testing for handheld dynamometry with a custom 3D-printed fixture.
Clear reporting of error, transparent constraints, and practical measurement protocols built to survive real-world clinical variability.
Output: Validated measurement approach and normative data, published and adopted for further clinical use.
Tools & output
Open data and research software I've released, my publications, and a side project.
Open data & research software
COMPWALK-ACL dataset
Multi-pace IMU gait kinematics from 92 participants: 25 healthy adults, 27 healthy adolescents, and 40 people with an ACL injury, 27 of them measured again three months after reconstruction. Slow, normal and fast walking. Open under CC BY 4.0; the trials are .mvnx files, which open in MVNX Viewer.
MVNX Viewer
A free browser viewer I built for .mvnx motion-capture files (Movella / Xsens MVN): 3D skeleton playback, joint-angle charts, foot contacts and centre of mass. Files are not uploaded.
Publications & side projects
25+ peer-reviewed publications
Research spanning wearable biomechanics, device validation, patient-reported outcomes, and population health. Editorial board member, BMJ Open Sport & Exercise Medicine.
Cuewand
A side project I built and maintain: a free phone remote for PowerPoint on Windows, with Next and Back, speaker notes per animation step, a talk timer (or none, for a lesson) and a laser pointer. Nothing to install on the phone.
Let's talk.
Open to industry and applied research roles. The quickest route is email; my CV is available on request.