Publications & Research

📄 View CV Return Home

Research portfolio

Decoding natural movement from wearable biosignals.

I build estimation and sequence models that recover movement from soft printed electrode arrays under unconstrained conditions—from finger motion and real-time gesture decoding to movement disorders.

High-resolution sEMG Motor control Sequence models Wearable interfaces

Publications

🎓 Google Scholar 🆔 ORCID

Peer-reviewed · newest first

Journal of Neural Engineering (2026)

Multimodal quantification of cognitive load using a printed wearable facial bio-potential system

Teitelbaum, D., Ibrahim, R., Man, H., Luxembourg, N., Re, D., & Hanein, Y.

Journal of Neural Engineering, Vol. 23, 046029.
DOI: 10.1088/1741-2552/ae89e9

🔗 View Paper

APL Bioengineering (2025)

Finger joint angle and gesture estimation under natural conditions with a soft printed electrode array

Luxembourg, N., Fekadu Marew, R., Teitelbaum, D., Ben-Dov, D., Siegelmann, H., & Hanein, Y.

APL Bioengineering, Vol. 9, No. 4, 046103. AIP Publishing.
DOI: 10.1063/5.0270645

🔗 View Paper

Journal of Visualized Experiments (2025)

Capturing dynamic finger gesturing with high-resolution surface electromyography and computer vision

Luxembourg, N., Ben-Dov, D., Fekadu Marew, R., Teitelbaum, D., VÄ—braitÄ— Adereth, I., & Hanein, Y.

Journal of Visualized Experiments (JoVE), No. 217, e67766.
DOI: 10.3791/67766

🔗 View Paper

IEEE EMBC (2025)

Finger joint angle and gesture estimation under dynamic hand position using a soft printed electrode array

Luxembourg, N., VÄ—braitÄ—, I., Siegelmann, H., & Hanein, Y.

47th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Copenhagen, Denmark, pp. 1–7.
DOI: 10.1109/EMBC58623.2025.11254337

🔗 View Paper 📊 View Poster

Manuscripts in preparation

In Preparation

Deconstructing motor signals: from variable-duration sEMG–IMU tokens to elementary actions and the syntax of natural movement

Luxembourg, N., Siegelmann, H., & Hanein, Y.

In Preparation

IMU-referenced decomposition of lower-limb sEMG using coupled non-negative matrix factorization: a proof-of-concept study in Parkinson's disease with and without freezing of gait

Luxembourg, N., Siegelmann, H., & Hanein, Y.

Conference presentations

  • • Poster — 47th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Copenhagen, Denmark, 2025.
  • • Talk — "Finger Joint Angle and Gesture Estimation in Dynamic Hand Postures Using a Soft Printed Electrode Array." IEEE International Conference on the Science of Electrical Engineering (ICSEE), Jerusalem, June 2026 (presented by a colleague); Physical Electronics Department Meetup, Tel Aviv University, February 2026.
  • • Poster — Karniel Computational Motor Control Workshop (KCMCW), Tel Aviv University, 2026.
  • • Poster — 14th Annual Nano Workshop, Jan Koum Center for Nanoscience and Nanotechnology, Tel Aviv University, 2026.

Education

Ph.D., Electrical and Computer Engineering

2024–present

Tel Aviv University

Supervisors: Prof. Yael Hanein (TAU) and Prof. Hava Siegelmann (UMass Amherst)

M.Sc., Brain Sciences (Computation & Information Processing)

2020–2024

The Hebrew University of Jerusalem

Advisor: Prof. Idan Segev · GPA 90.4 · Thesis 92.5 · Final examination 94

B.Sc., Computer Science (Minor: Computational Neuroscience)

2017–2020

The Hebrew University of Jerusalem

GPA 90.3 · Class rank 92 / 369

Featured Work

📊

Research Poster

Finger Joint Angle Estimation

Research on finger joint angle and gesture estimation under dynamic hand position using soft printed electrode arrays.

EMG Soft Electrodes Joint Estimation
View Poster 📄 Download PDF
🎓

MSc Thesis

Single Neuron Functional Analysis

MSc thesis using information theoretical approach to analyze computational capabilities of individual neurons and their input-output relationships.

MSc Research Information Theory Neuronal Modeling
Read Thesis 📄 Download PDF
🧩

Movement Syntax

Deconstructing Motor Signals

Sequence models that segment continuous sEMG–IMU recordings into variable-duration tokens, testing whether natural movement decomposes into a reusable vocabulary of elementary actions.

Sequence Models sEMG–IMU Tokenization
Manuscript in preparation

Research Experience

Naturalistic Motor Control Using High-Resolution sEMG

2024–present

Tel Aviv University

  • • Built and validated an estimation pipeline that recovers finger joint angles and discrete gestures from soft printed electrode array recordings under dynamic hand position, benchmarked against computer-vision hand tracking.
  • • Developing sequence models that segment continuous sEMG–IMU recordings into variable-duration tokens, testing whether natural movement decomposes into a reusable vocabulary of elementary actions; manuscript in preparation.
  • • Building a real-time decoding pipeline that carries this offline estimation work into online operation, for low-latency gesture-driven interaction; early stage.
  • • Developing an IMU-referenced decomposition algorithm for sEMG based on coupled non-negative matrix factorization, validated on lower-limb recordings in Parkinson's disease with and without freezing of gait; manuscript in preparation.

Single-Neuron Functional Analysis: An Information-Theoretic Approach

2021–2024

Segev Lab, The Hebrew University of Jerusalem

  • • Showed that neurons are not computationally interchangeable: human L2/3 pyramidal cells carry measurably more information than rat L5b, dendritic morphology accounts for part of the gap, and removing NMDA nonlinearity collapses complexity in every model — evidence that single-cell biophysics, not only circuit wiring, sets a neuron's computational capacity.
  • • Built the measures that make such comparisons possible — a firing-rate-invariant entropy and statistical-complexity measure of output spike trains, plus an objective network-depth benchmark replacing subjective model comparison — letting any neuron model be placed on a common complexity scale.
🎓 Read the Thesis

Neural Decoding of Visual Attention from EEG

2019–2020

The Hebrew University of Jerusalem

  • • Applied machine-learning classifiers to EEG recordings to decode covert visual attention; undergraduate research project, exploratory.

Teaching

Teaching Assistant — Dynamical Systems and the Neuron

The Hebrew University of Jerusalem · 2021–2022

Teaching Assistant — Brain–Computer Interfaces as Assistive Technology

The Hebrew University of Jerusalem · 2021–2022

Technical Skills

Programming

Python C C++ Java

Machine Learning

PyTorch Deep Learning Transformers Temporal Convolutional Models Autoregressive Sequence Models Non-Negative Matrix Factorization

Signals & Instrumentation

Surface EMG EEG IMU High-Density Soft Electrode Arrays Arduino / Embedded Prototyping Computer-Vision Motion Capture

Interactive Systems

Unity Real-Time Gesture-Driven Interfaces Low-Latency Decoding Pipelines

Domain

Motor Control Computational Neuroscience Brain–Computer Interfaces Information Theory / Entropy Estimation Biophysical Compartmental Modeling