NeuroAI Lab     

Welcome to the NeuroAI Lab at the Wyss Center for Bio and Neuroengineering. We develop machine learning methods and data systems for neuroinformatics and medical AI, with a focus on longitudinal neural, wearable, and clinical time-series data. Our goal is to understand human state, intention, and behavior from complex biological signals, and to transform these signals into reliable tools for neuroscience, medicine, rehabilitation, and human-machine interaction.

A central theme of our work is modeling human state from neural and biomedical signals as they change over time. We study latent representations in high-dimensional neural dynamics, develop foundation models for neural and biomedical time series, and build real-time decoding systems for closed-loop applications. Our current applications include neurorehabilitation and assistive systems for Parkinson’s disease, spinal cord injury, and stroke. We see neuroinformatics as a bridge between AI, neuroscience, medicine, and human-centered systems. By developing reliable models for longitudinal neural and biomedical data, we aim to support new tools for neurorehabilitation, neurological disorders, aging research, assistive technologies, and data-driven medical discovery.

Research topics

• Longitudinal Neuroinformatics
Brain and biomedical signals change across days, months, people, devices, and clinical conditions. We develop machine learning methods for modeling non-stationary neural and biological time series, including neural drift, cross-day alignment, cross-subject transfer, and uncertainty-aware decoding.

• Real-Time Medical AI and Closed-Loop Systems
We build low-latency AI systems that can operate in real time with humans in the loop. Applications include adaptive neurotechnology, assistive wearables, and neurorehabilitation.

• Multimodal Human-State Modeling
We combine neural, behavioral, wearable, physiological, and clinical data to infer human state, intention, and functional ability. This line of work connects neuroinformatics with human-centered AI, rehabilitation, aging research, and adaptive human-machine interaction.

Open Tools for Reproducible Neuroinformatics

We are committed to open science and reproducibility. We develop and maintain open-source tools for neuroinformatics, including machine learning libraries and analysis pipelines. Our tools are designed to be scalable, and adaptable to a wide range of neuroscience and biomedical applications. We also share our models to facilitate collaboration and accelerate progress in the field.

Collaboration projects

• Restoring communication pathways after spinal cord injury or stroke
   PI: Dr. Kyuhwa Lee, Wyss Center (CH) and Dr. Yun-Jae Won, KETI (KR)
The aim of the project is to develop high-performance brain signal decoding algorithms that can deal with neural drift, coupled with a soft & lightweight exosuit to support arm and hand movement. This international project is supported by bilateral funding from Switzerland's Innosuisse and South Korea's KIAT, bringing together expertise from both countries. Its significance lies in its potential to accelerate neuro-rehabilitation for people with stroke and spinal cord injuries by using novel peripheral nerve stimulation exosuit design that allows for sustained postures without fatigue.

• Connection between the gut and the brain
   PI: Dr. Michalina Gora, Wyss Center
We explore the critical connection between the gut and the brain. The aim is to understand the enteric nervous system and its role in health, with the goal of finding new biomarkers and therapies for brain disorders like Parkinson's, dementia, and depression. This research is vital due to the increasing recognition of the gut's influence on brain function and the current lack of tools to monitor this complex interplay.

• Lightsheet microscopy brain segmentation
   PI: Dr. Stéphane Pagès, Wyss Center
This project focuses on developing and deploying AI-based image analysis pipelines for the ALICe lightsheet microscopy platform at the Wyss Center. The primary goal is to create robust algorithms for brain registration, volumetric segmentation, and image stitching.

• Transcranial temporal interference stimulation
   PI: Prof. Friedhelm Hummel, EPFL
This Lighthouse project aims to use AI to rapidly and accurately predict electric fields induced by brain stimulation, overcoming the computational challenges and expertise requirements of current methods such as Finite Element Method. The goal is to enable personalised and optimised stimulation parameters for individual patients to integrate neuromodulation into clinical practice for various neurological and psychiatric conditions.