Dr. Sc. · ETH Zürich
Computational Biomechanics Researcher
I turn raw, messy wearable and health data into clear, validated insight at scale — combining statistics and modern machine learning to understand how people move.
I turn raw, messy wearable and health data into clear, validated insight at scale — combining statistics and modern machine learning to understand how people move.
I'm a computational biomechanics researcher who recently completed a doctorate (Dr. Sc.) at ETH Zürich, in the Laboratory for Movement Biomechanics under Prof. William Taylor and Dr. Navrag Singh. My work sits at the intersection of engineering, biomechanics, wearable sensing, machine learning, and clinical translation — developing objective measures of human movement that work beyond specialised laboratories.
In practice, I turn raw, messy wearable and health data into clear, validated measures of movement — across small, controlled laboratory experiments, a 2,000+ participant ageing cohort, and real-world wearable studies. An engineering foundation (MEng, UCL) underpins how I move fluently between signal processing, statistics, physiology, and computation, rather than staying inside any single discipline.
I reason from first principles and pair modern tools — including AI and LLMs — with scientific judgement: I use them to accelerate exploration, analysis, and writing, while relying on domain reasoning to know when to trust a result and when not to. I'm at my most useful on open-ended, interdisciplinary problems, where the right questions emerge through building and iteration.
Ultimately, I'm motivated by problems where engineering, AI, and clinical science meet, and where rigorous analysis leads to practical improvements in how we understand human health and movement. Outside work, I enjoy skiing, wake surfing, handstand training, tennis, and travel.
8 published journal articles + 2 under review + 1 preprint · full list on Google Scholar
Deep Kinematics: uncovering human biomechanics with IMUs and deep learning Under review
IEEE Journal of Biomedical and Health Informatics
Footprints of hearing loss: can gait features provide clues for identifying deficits in hearing? Under review
IEEE Access
Leveraging deep learning and wearables for automatically identifying gait events: effects of age and sensor location on gait-event assessment
IEEE Sensors DOI ↗
Adaptive gait responses to varying weight-bearing conditions: inferences from gait dynamics and H-reflex magnitude
Experimental Physiology DOI ↗
A deep-learning approach for automatically detecting gait events based on foot-marker kinematics in children with cerebral palsy — which markers work best for which gait patterns?
PLOS ONE DOI ↗
Walking in the free world: establishing normative trajectories for ecological assessment of robust gait variability with age Preprint
Preprint · in review DOI ↗
Sheep treadmill and real-life walking kinematic analysis using a novel wearable-based method
Frontiers in Animal Science DOI ↗
The effects of weight-bearing manipulations on gait and its underlying neural control mechanisms in toe-walking children
Frontiers in Human Neuroscience DOI ↗
Probing gait adaptations: the impact of aging on dynamic stability and reflex control mechanisms under varied weight-bearing conditions
European Journal of Applied Physiology DOI ↗
Classification of inertial sensor-based gait patterns of orthopaedic conditions using machine learning: a pilot study
Journal of Orthopaedic Research DOI ↗
How the CYBATHLON competition has advanced assistive technologies
Annual Review of Control, Robotics and Autonomous Systems DOI ↗
Age group identification using machine learning and IMU: a comparison of sensor placements Poster
ESMAC (European Society for Movement Analysis in Adults and Children), 2023
Deep Kinematics: uncovering human biomechanics with IMUs and deep learning — preliminary results Oral
International Society of Biomechanics (ISB), 2023
Automated gait-event detection using wearables/IMU for data acquisition and deep learning for placement classification Oral
International Society of Biomechanics (ISB), 2023
Towards automated gait-event detection using machine learning — what foot markers work best for what gait patterns? Oral
International Society of Biomechanics (ISB), 2021
How does modulating load impact the limits of stability during walking? Inferences from simulated body-weight support and load carriage Poster
International Society of Biomechanics (ISB), 2021
From laboratory to real-world gait: leveraging progressive transfer learning for accurate and reliable event detection
ACM Conference on Bioinformatics, Computational Biology & Health Informatics (BCB ’26), 2026 DOI ↗
Enhancing robustness using transfer learning: deep-learning-based gait-event detection for older adults Oral
European Society for Biomechanics (ESB), 2025
Does arm swing associate with variability among older individuals? Oral
European Society for Biomechanics (ESB), 2024
Investigating the effect of auditory noise on gait stability in young and elderly healthy individuals Poster
ESMAC (European Society for Movement Analysis in Adults and Children), 2023
Distinguishing healthy subjects from patients with different degenerative diseases based on gait pattern — a machine learning approach Poster
Gait & Posture, Vol. 100 Suppl. 1, 2023, S74–S75
Soleus H-reflex gain under different bodyweight conditions during walking in children, young and older adults Poster
Gait & Posture, Vol. 97 Suppl. 1, 2022, S74–S75 DOI ↗
Head of Discipline — FES Bike Race
CYBATHLON
Defined competition standards for assistive technologies and coordinated international teams and stakeholders across global hubs.
Research Assistant
ETH Zürich
Investigated neuromuscular control of gait stability using H-reflex measurements under varying weight-bearing conditions.
Research Intern
Hanyang University
Developed IMU-based algorithms for gait-cycle detection and feature extraction to support biometric gait recognition.
Part-time Technology Consultant
Samsung Advanced Institute of Technology
Evaluated emerging technologies and patent landscapes, delivering time-critical insights that supported adoption decisions.