TUM Neurotechnology Research Alliance
Research
The alliance studies neural signals, develops interfaces between the nervous system and technology, and moves viable approaches into clinical-translational programs.
Four closely interlinked research areas structure this work
1 - Sensorimotor rehabilitation
Neural interfaces and closed-loop systems to support or restore movement functions.
Cortical signals are linked to technical assistance systems and neuroprostheses, and in the longer term also to spinal cord stimulation. Applications are investigated in traumatic spinal cord injury, after stroke, and in other severe motor impairments.
The projects combine intracortical recording, AI-supported decoding, and electrophysiological methods with clinically defined rehabilitation goals. In the long term, the aim is closed-loop systems that detect a movement intention and translate it directly into targeted stimulation or assistive movement.
Focus areas
The area covers intracortical BCIs for motor functions, brain-spine interfaces, and spinal cord stimulation. Additional work addresses high-resolution EMG and motor neuron decoding, closed-loop architectures, and clinical evaluation over the course of rehabilitation.
2 - Communication and cognition
How are communicative and higher cognitive functions represented in the human brain, and how can neural signals be used for assistive applications?
This axis works with invasive human neurophysiology, high-resolution microelectrode recordings, and patient-specific decoding. Its scope reaches from the analysis of individual nerve cells and neural populations to brain-computer interfaces for people whose ability to communicate is severely impaired.
Clinically, the main focus is loss of language after stroke. In parallel, basic science projects examine higher cognitive functions and develop the methodology for BCI applications that go beyond simple motor control.
Focus areas
Work covers acute and chronic intracortical recordings, single-cell and population analyses, and the decoding of speech and communication. Further topics are the representation of higher cognitive functions, patient-specific BCI models, and functional mapping for preoperative target planning.
Associated projects: Acute implantation platform, RHETORICAL, MEMCIRCUIT, DFG Vocal Communication, Prefrontal Flexibility
3 - Robotic assistance and neuroprosthetics
Wearable robotic systems and neuroprostheses for people with neurologically caused movement disorders.
The engineering work combines biomechanical models, sensor systems and control engineering with compliant actuation and sensorimotor human-robot interaction. Topics include soft exosuits for the lower limb, active orthoses, hand neuroprosthetics, and virtual embodiment paradigms.
The systems are meant to fit as unobtrusively as possible into the movement and everyday life of their users. Clinical testing and individual adjustment are therefore part of the development from the outset.
Focus areas
This includes soft wearable robotics and exosuits, active orthoses, and hand neuroprosthetics. The decoding of grasping and manipulation movements, myoelectric control, and virtual reality for embodiment studies complete the field.
Associated projects: Soft Wearable Robotics, Hand neuroprosthetics, AID 2.0, nTMS-VR-Embodiment
4 - Ethics and governance for data and AI
Methodological, technical, ethical and regulatory foundations for all research and development programs.
Neural interfaces produce high-dimensional and particularly sensitive data, and they have to adapt to individual users in real time. This creates a need for patient-specific decoding methods, for energy-efficient neuromorphic hardware and for a data and computing infrastructure designed to match.
Neuroethics, data protection, neurorights, clinical trial planning, and medical device law are considered from the first project outline onwards. Governance is therefore part of the research architecture itself.
Focus areas
Current topics are machine learning for neural data, neuromorphic signal processing and hardware, and real-time and edge decoding. On the regulatory side there are data infrastructure and data protection, neuroethics and neurorights, and the planning of clinical trials.
Associated projects: Neuromorphic AI (AID 2.0), Patient-specific decoding, Governance framework
Shared translational platform
We start in the neurosurgical operating room at TUM University Hospital — where clinical reality becomes a platform for discovery.
From there, the alliance brings together disciplines that have rarely been combined in this way: neurosurgery, neuroscience, engineering, robotics, artificial intelligence, and neuroethics.
By integrating clinical needs, scientific measurement, and technological development under real-world conditions, we aim to move seamlessly from clinical insight to innovation.