PhD researcher · ETH Zurich

Giusy Spacone

I design wearable, multimodal neuromotor interfaces, focusing on efficient algorithms that run on the device and on user-centred applications.

Deep learningEdge AIBiosignal processingWearablesHuman–machine interfacesSensor fusionEmbedded firmware
I am a PhD researcher at the Integrated Systems Laboratory at ETH Zurich. My research focuses on next-generation wearable, multimodal neuromotor interfaces for novel human–machine interaction strategies. Bringing together surface electromyography (sEMG), A-mode ultrasound and inertial (IMU) sensing, I work across the full pipeline: from wearable form-factor development and system integration to machine learning for physiological signal decoding, on-device deployment and validation through human-subject studies, towards end-to-end, user-centred solutions for real-world use cases.
Portrait of Giusy Spacone

About me

Background

Education

Jun 2024 – exp. Sep 2028

PhD, Information Technology & Electrical Engineering (D-ITET)

ETH Zurich, Integrated Systems Laboratory

Professor: Prof. Luca Benini · Co-advisor: Dr. Andrea Cossettini.

2021 – 2024

MSc, Biomedical Engineering (Bioelectronics & Biosensors)

ETH Zurich

GPA 5.74 / 6.0. Thesis: armband for concurrent ultrasound–EMG acquisition to enable new control strategies for dynamic arm movements. Projects on real-time gesture classification on STM32 (Cortex-M4), spiking neural networks on the Dynap-SE neuromorphic processor (Institute of Neuroinformatics, UZH/ETH Zurich), and a perfusion-system driver for neuronal cell cultures.

2018 – 2021

BSc, Biomedical Engineering

Politecnico di Milano

110 / 110 cum laude. Thesis: GSR signal analysis for human–machine interaction (NearLab).

Experience

Jun 2024 – Present

PhD Candidate – Wearable Neuromotor Interfaces & Edge AI

ETH Zurich, Integrated Systems Laboratory

Wearable, multimodal neuromotor interfaces through hardware–software co-design, on the open-source, ultra-low-power BioGAP-Ultra and WULPUS platforms. See the research pillars below.

Feb 2023 – Jul 2023

Rotational Intern – Electronics, R&D and Data Science

Nanoleq

Rotated across the electronics, R&D and data-science teams: PCB design and firmware for automated lithium-battery life-cycle testing with data-analysis pipelines; defined experimental protocols for ESD characterisation of wearable biosensing textiles; designed and ran a multimodal stress-monitoring study (ECG, respiration, EDA), from protocol definition to data analysis.

Skills

Machine learning & signal processing

  • Languages & frameworks: Python, PyTorch, scikit-learn, SciPy
  • Expertise: physiological signal processing, time-series classification and regression, multimodal sensor fusion, CNNs and transformer-based sequence models, fine-tuning for cross-day recalibration

Embedded systems

  • Embedded AI: ARM Cortex-M (nRF52), RISC-V (GAP9); LiteRT / TensorFlow Lite and research deployment frameworks, INT8 quantisation, embedded inference, power profiling
  • Embedded firmware: C, Zephyr RTOS, ESP-IDF
  • PCB design: KiCad, Altium

Wearable sensing & experimental research

  • Wearable sensing: sEMG, A-mode US, IMU; wearable system design, sensor integration
  • Experimental research: human-subject study design, ethics applications

Languages

  • ItalianNative
  • EnglishC1
  • GermanB1

Soft skills

Leadership & team buildingMentoring (15+ supervised student projects)Communication: workshops, talks, journalismResilience & time management (former competitive alpine skier, now running & cross-training)

Research

Key research pillars

I combine surface electromyography (sEMG), A-mode ultrasound and inertial sensing, with two main application domains: hand-gesture recognition with continuous kinematic tracking, and silent-speech interfaces.

Click a card to see its publications.

Illustration: Wearable hand & arm tracking

Wearable hand & arm tracking

Best Conference Paper · IEEE SMC 2026 A-mode ultrasoundsEMGIMU

Continuous hand–wrist tracking is key to enabling natural human–machine interaction, yet challenges remain to ensure robustness across users and sensor repositioning. This research combines established modalities (sEMG, IMU) with A-mode ultrasound to capture complementary muscle information for robust, low-power kinematic tracking and gesture recognition, towards reliable real-time interaction across multiple days.

  • Wearable A-mode US for kinematic tracking — designed a four-transducer US armband interfaced with the WULPUS acquisition system; integrated motion capture and an instrumented glove for ground truth; led the data collections; developed XGBoost and CNN regression models, INT8-quantized and deployed on nRF52. 3-DoF: 5 subjects over 9 sessions, inter-session RMSE 11.1°. 23-DoF: 11k-parameter CNN, 29.1 ms inference on the nRF52 MCU, < 35 mW for sensing, inference and BLE result transmission.
  • Concurrent EMG + US fusion — designed a flexible EMG PCB and wearable armband interfacing with the BioGAP acquisition system; implemented firmware-level synchronisation for concurrent EMG and US acquisition; designed and led data collections; developed a CNN-based EMG–US late-fusion architecture. 23-DoF: inter-session RMSE 10.6° vs. 12.0° (EMG only) and 13.1° (US only).
  • US + IMU for real-time VR — led multi-day data collections, developed late-fusion CNNs with personalised recalibration, and built the real-time pipeline into a Unity VR environment. 5 subjects over 5 days: 80% / 77% inter-session accuracy; 88–96% online VR success after 5 min of recalibration.

Now: distributed EMG + US + IMU fusion for VR and robotic control.

Demo videoWULPUSBioGAP Publications (5)
Illustration: Wearable silent-speech interfaces

Wearable silent-speech interfaces

sEMGSilent speechEdge AI

Silent-speech interfaces decode speech-related intent from physiological signals, enabling private and accessible communication, but existing approaches rely on obtrusive sensing and struggle to stay robust across sessions. This research explores unobtrusive, wearable interfaces with on-device processing.

  • Contributions — conceptualised SilentWear, a fully wearable, dry-electrode textile sEMG neckband built around the BioGAP platform; conceived and ran multi-day, multi-subject studies with vocalised and silent speech; developed SpeechNet, a compact 15k-parameter CNN for edge deployment, with a personalised fine-tuning strategy for inter-session robustness.
  • Results — 8-command recognition across 4 subjects: 84.8% / 77.5% accuracy (vocalised / silent, aggregated evaluation); fine-tuning recovers > 10 percentage points of accuracy lost to repositioning within < 10 min of personalisation; 2.47 ms inference at the edge (GAP9).

Now: EMG, EEG and audio fusion for continuous speech recognition.

CodeDataset Publications (2)
Illustration: Foundation models for the edge

Foundation models for the edge

Foundation modelssEMGTinyML

Foundation models can learn transferable representations from physiological signals, reducing the need for task-specific models and labelled data. However, variability across acquisition systems, subjects and tasks challenges generalisation, and computational cost limits wearable deployment. This research investigates compact foundation models that bridge transferable representation learning and edge inference.

  • Contributions — conceived TinyMyo, the first EMG foundation model at the edge, and defined its research direction: dataset unification, the pre-training framework and multi-task downstream evaluation; supervised its development and experimental validation within a Master’s thesis.
  • Results — a 3.6M-parameter transformer backbone evaluated on gesture recognition (NinaPro DB5, UCI-EMG, EPN-612, Generic Neuromotor Interface by Meta), kinematic regression (NinaPro DB8) and silent speech (Gaddy). Deployed on GAP9, where a 1.9M-parameter variant achieves 89 ms inference on 1-s windows.

Now: foundation models for A-mode ultrasound at the edge.

Code Publications (1)
Illustration: Wireless body-area network

Wireless body-area network

In progress EMGUltrasoundIMU

I am developing a distributed body-area network built from compact, multimodal sensing nodes with EMG, ultrasound and IMU sensing. A centralised computing “brain” coordinates the nodes, so the same wearable network can run several applications at once.

Publications (1)

Research

Selected publications

A Wearable Multimodal Ultrasound + Inertial System for Real-Time Virtual Reality Interaction

G. Spacone, S. Frey, E. Baraldi, M. Orlandi, L. Benini, A. Cossettini

2026 ·IEEE International Conference on Systems, Man, and Cybernetics (SMC) Best Conference Paper Award Demo videoOpen access

Toward Whole Hand and Wrist Kinematic Tracking with a Wearable A-mode Ultrasound Probe

G. Spacone, L. Benini, A. Cossettini

2026 ·IEEE Open Journal of Ultrasonics, Ferroelectrics, and Frequency Control IEEEOpen access

SilentWear: An Ultra-Low-Power Wearable System for EMG-based Silent Speech Recognition

G. Spacone, S. Frey, G. Pollo, A. Burrello, D. Jahier Pagliari, V. Kartsch, A. Cossettini, L. Benini

2026 ·IEEE Sensors Journal IEEEOpen accessCodeDataset

TinyMyo: A Tiny Foundation Model for Flexible EMG Signal Processing at the Edge

M. Fasulo, G. Spacone, T. M. Ingolfsson, Y. Li, L. Benini, A. Cossettini

2026 ·IEEE Transactions on Biomedical Engineering IEEEOpen accessCode

Tracking of Wrist and Hand Kinematics with Ultra-Low-Power Wearable A-mode Ultrasound

G. Spacone, S. Vostrikov, V. Kartsch, S. Benatti, L. Benini, A. Cossettini

2024 ·IEEE Transactions on Biomedical Circuits and Systems IEEE

Full list on Google Scholar.

Research

Academic service & teaching

Teaching & supervision

  • Primary supervisor — 15+ student research projects, defining research directions, ETH Zurich (2024 – present)
  • Teaching assistant — Wearable Ultrasound (P&S), ETH Zurich (2024 – present)
  • Teaching assistant — Clinical & Movement Biomechanics, ETH Zurich (2022 – 2024)

Academic service

  • Workshop instructor — A-mode US at the edge: a hands-on introduction to deployment with open-source tools, IEEE Central and Eastern European Ultrasonics Symposium, Warsaw (2026)
  • Journal reviewer — IEEE Transactions on Human-Machine Systems; Nature Communications
  • Conference reviewer — NeurIPS workshop

Awards

2026

Best Conference Paper Award

IEEE International Conference on Systems, Man, and Cybernetics (SMC)

For "A wearable multimodal ultrasound + inertial system for real-time virtual reality interaction".

2019

Best Freshman Prize

Politecnico di Milano

Awarded to the top first-year students.

2018

Alfiere del Lavoro

President of the Italian Republic

Honor from the President of the Italian Republic (25 students nationwide).

Beyond research

Community & leadership

I believe a better society is built by people who take an active part in it. Social and community work is a big part of my life: I founded and lead the Swiss Italian Society in Zurich, and I care about giving back by sharing my experience and expertise with younger generations.

2023 – present

Founder & President, Swiss Italian Society (SIS)

The association connecting young Italians who live, work and study in Zurich.

Visit sis-zh.com

2024 – present

Head of External & Institutional Relations

United Italian Society – Switzerland

2020 – 2024

Lead the Future mentee · ETH Ambassador

ETH Zurich

2018 – 2021

Student representative, Biomedical Engineering

Politecnico di Milano

Outside work

Alpine skiing (former competitive athlete)Running & cross trainingHiking & exploringWriting & science communication

Contact

Get in touch

Happy to talk about wearables, embedded machine learning and human–machine interfaces.