Felix Christian Bauer

Felix Christian Bauer

Auto-generated profile, based on publicly available information

Felix Christian Bauer

SynSense, Zürich, Switzerland

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 November 2019 in IEEE Transactions on Biomedical Circuits and Systems
Verified
Journal Article
Closed access
FBFelix Christian Bauer
Felix Christian Bauer
DMDylan Richard Muir
Dylan Richard Muir
GIGiacomo Indiveri
Giacomo Indiveri
Accurate detection of pathological conditions in human subjects can be achieved through off-line analysis of recorded biological signals such as electrocardiograms (ECGs). However, human diagnosis is time-consuming and expensive, as it requires the time of medical professionals. This is especially inefficient when indicative patterns in the biological signals are infrequent. Moreover, patients with suspected pathologies are often monitored for extended periods, requiring the storage and examination of large amounts of non-pathological data, and entailing a difficult visual search task for diagnosing professionals. In this work we propose a compact and sub-mW low power neural processing system that can be used to perform on-line and real-time preliminary diagnosis of pathological conditions, to raise warnings for the existence of possible pathological conditions, or to trigger an off-line data recording system for further analysis by a medical professional. We apply the system to real-time classification of ECG data for distinguishing between healthy heartbeats and pathological rhythms. Multi-channel analog ECG traces are encoded as asynchronous streams of binary events and processed using a spiking recurrent neural network operated in a reservoir computing paradigm. An event-driven neuron output layer is then trained to recognize one of several pathologies. Finally, the filtered activity of this output layer is used to generate a binary trigger signal indicating the presence or absence of a pathological pattern. We validate the approach proposed using a Dynamic Neuromorphic Asynchronous Processor (DYNAP) chip, implemented using a standard 180 nm CMOS VLSI process, and present experimental results measured from the chip.

16 September 2024 in Nature Communications
Verified
Journal Article
Open access
PDF file
JPJens E. Pedersen
Jens E. Pedersen
SASteven Abreu
Steven Abreu
MJMatthias Jobst
Matthias Jobst
GLGregor Lenz
Gregor Lenz
VFVittorio Fra
Vittorio Fra
JEJason K. Eshraghian
Jason K. Eshraghian
Spiking neural networks and neuromorphic hardware platforms that simulate neuronal dynamics are getting wide attention and are being applied to many relevant problems using Machine Learning. Despite a well-established mathematical foundation for neural dynamics, there exists numerous software and hardware solutions and stacks whose variability makes it difficult to reproduce findings. Here, we establish a common reference frame for computations in digital neuromorphic systems, titled Neuromorphic Intermediate Representation (NIR). NIR defines a set of computational and composable model primitives as hybrid systems combining continuous-time dynamics and discrete events. By abstracting away assumptions around discretization and hardware constraints, NIR faithfully captures the computational model, while bridging differences between the evaluated implementation and the underlying mathematical formalism. NIR supports an unprecedented number of neuromorphic systems, which we demonstrate by reproducing three spiking neural network models of different complexity across 7 neuromorphic simulators and 4 digital hardware platforms. NIR decouples the development of neuromorphic hardware and software, enabling interoperability between platforms and improving accessibility to multiple neuromorphic technologies. We believe that NIR is a key next step in brain-inspired hardware-software co-evolution, enabling research towards the implementation of energy efficient computational principles of nervous systems. NIR is available atneuroir.org

22 October 2019 in IEEE Transactions on Robotics
Verified
Journal Article
Closed access
Luca ToninLuca Tonin
Luca Tonin
FBFelix Christian Bauer
Felix Christian Bauer
JMJosé del R. Millán
José del R. Millán
Despite the growing interest in brain-machine interface (BMI)-driven neuroprostheses, the translation of the BMI output into a suitable control signal for the robotic device is often neglected. In this article, we propose a novel control approach based on dynamical systems that was explicitly designed to take into account the nature of the BMI output that actively supports the user in delivering real-valued commands to the device and, at the same time, reduces the false positive rate. We hypothesize that such a control framework would allow users to continuously drive a mobile robot and it would enhance the navigation performance. 13 healthy users evaluated the system during three experimental sessions. Users exploit a 2-class motor imagery BMI to drive the robot to five targets in two experimental conditions: with a discrete control strategy, traditionally exploited in the BMI field, and with the novel continuous control framework developed herein. Experimental results show that the new approach: 1) allows users to continuously drive the mobile robot via BMI; 2) leads to significant improvements in the navigation performance; and 3) promotes a better coupling between user and robot. These results highlight the importance of designing a suitable control framework to improve the performance and the reliability of BMI-driven neurorobotic devices.

13 March 2026
Verified
Conference Paper
Open access
AKAkash Koppa
Akash Koppa
SPSon Pham-Ba
Son Pham-Ba
FBFelix Bauer
Felix Bauer
OBOlivier Bonte
Olivier Bonte
OBOscar Baez-Villanueva
Oscar Baez-Villanueva
SBSara Bonetti
Sara Bonetti
Hybrid modeling, which integrates physics-based and machine learning (ML) components, is a growing research area in hydrology and the broader Earth Science community. By combining the interpretability of process-based models with the predictive power of data-driven algorithms, these hybrid architectures offer improved accuracy and representation of complex environmental processes. However, their adoption is currently constrained by significant challenges regarding FAIR principles (Findable, Accessible, Interoperable, Reusable) . Unlike traditional physics-based models, the reusability of hybrid systems is frequently hindered by the dynamic nature of ML components, which are inextricably linked to specific training datasets and hyperparameter configurations. Furthermore, existing data data and model repositories are rarely designed to host such models.To address these systemic barriers, we collaboratively designed and implemented a standardized FAIR protocol specifically tailored for hydrological hybrid models. This framework, termed as FRAME, consists of three critical components: (a) a set of interoperability coding standards for the physics and ML modules, (b) a unified metadata specification that captures the disparate requirements of both physics-based parameters and ML architectures, and (c) a specialized online repository designed for the persistent hosting and sharing of integrated hybrid assets. To facilitate user adoption, we developed an associated command line interface (CLI) for automated retrieval and setup of these models. To ensure the long-term impact and scalability of this protocol, we are actively soliciting participation from the global hydrologic modeling community. By establishing a community-driven standard, this protocol aims to provide a robust foundation for the transparent, reproducible, and collaborative advancement of hybrid modeling in hydrology.

13 November 2019 in arXiv
Verified
Preprint
Open access
PDF file
FBFelix Christian Bauer
Felix Christian Bauer
DMDylan Richard Muir
Dylan Richard Muir
GIGiacomo Indiveri
Giacomo Indiveri
Accurate detection of pathological conditions in human subjects can be achieved through off-line analysis of recorded biological signals such as electrocardiograms (ECGs). However, human diagnosis is time-consuming and expensive, as it requires the time of medical professionals. This is especially inefficient when indicative patterns in the biological signals are infrequent. Moreover, patients with suspected pathologies are often monitored for extended periods, requiring the storage and examination of large amounts of non-pathological data, and entailing a difficult visual search task for diagnosing professionals. In this work we propose a compact and sub-mW low power neural processing system that can be used to perform on-line and real-time preliminary diagnosis of pathological conditions, to raise warnings for the existence of possible pathological conditions, or to trigger an off-line data recording system for further analysis by a medical professional. We apply the system to real-time classification of ECG data for distinguishing between healthy heartbeats and pathological rhythms. Multi-channel analog ECG traces are encoded as asynchronous streams of binary events and processed using a spiking recurrent neural network operated in a reservoir computing paradigm. An event-driven neuron output layer is then trained to recognize one of several pathologies. Finally, the filtered activity of this output layer is used to generate a binary trigger signal indicating the presence or absence of a pathological pattern. We validate the approach proposed using a Dynamic Neuromorphic Asynchronous Processor (DYNAP) chip, implemented using a standard 180 nm CMOS VLSI process, and present experimental results measured from the chip.