NTNicolas Thomé

Nicolas Thomé

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Nicolas Thomé has a total of 6 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Origin of frequency clusters and self-organized triplet locking in the Kuramoto model with inertia

    by Yannick Schöhs, Nicolas Thomé, Katharina Krischer

    We investigate the origin of frequency clusters – states in which multiple groups of oscillators with different mean frequencies coexist – in the globally coupled Kuramoto model with inertia and identical oscillators. Two frequency clusters are studied in the thermodynamic limit, three frequency clusters with a system of seven oscillators. Using bifurcation analysis, we demonstrate that in both cases the frequency clusters emerge through homoclinic bifurcations. In the case of three frequency clusters, this necessarily entails the formation of a triplet locked state, characterized by rational relations among the mean frequency differences. The individual clusters may lose phase-synchrony via either transcritical or period-doubling bifurcations. Finally, we establish that Hopf bifurcations cannot generate frequency clusters in phase oscillator systems. Instead, they can only arise through global bifurcations.

  • Open Access

    Synchronized Bimodal Amplitude Patterns in Heterogeneous Oscillatory Media - Experiment and Theory

    by Nicolas Thomé, Yukiteru Murakami, Katharina Krischer

    We study an intricate mechanism of pattern formation in globally coupled heterogeneous oscillatory media. In anodic electrochemical etching of silicon, the electrode surface splits into two amplitude-phase regions, while all oscillators remain frequency-locked. Additionally, the relative ratio of the pattern can be tuned via a coupling term. We introduce a heterogeneous, complex Ginzburg-Landau equation with global coupling to reproduce these patterns and study their formation. Neglecting diffusion shows that frequency entrainment arises from amplitude adaptation. Diffusion, on the other hand, enforces the selection of a unique cluster ratio. In quantitative agreement with simulations, a center-manifold reduction yields a Lyapunov functional that predicts the selected ratio. Both of these results establish a theoretical framework that connects experiment and theory. Moreover, they show how heterogeneity and diffusion convert degenerate cluster dynamics into robust pattern selection.


  • Hierarchical clustering in mean-field coupled Stuart–Landau oscillators

    by Nicolas Thomé, Matthias Wolfrum, Katharina Krischer

    Clustered solutions in oscillator networks provide an important insight into how a system might diversify from a synchronous solution into spatiotemporal complex solutions. They can, therefore, form a link between the fully synchronized and incoherent states. Despite their fundamental role in coupled oscillator dynamics, our understanding of how these clusters form and differentiate is still quite limited. Here, we study an ensemble of globally coupled Stuart–Landau oscillators and focus on how 3-cluster solutions emerge from 2-cluster solutions and how the different 3-cluster solutions are organized in parameter space. We show that the arrangement of the clusters is dictated by a codimension-two point, which we call a Type-II cluster singularity. Furthermore, our study points to a hierarchical structure of multi-cluster solutions.


  • 3D spatial priors for semi-supervised organ segmentation with deep convolutional neural networks

    by Olivier Petit, Nicolas Thome, Luc Soler

    Purpose Fully Convolutional neural Networks (FCNs) are the most popular models for medical image segmentation. However, they do not explicitly integrate spatial organ positions, which can be crucial for proper labeling in challenging contexts. Methods In this work, we propose a method that combines a model representing prior probabilities of an organ position in 3D with visual FCN predictions by means of a generalized prior-driven prediction function. The prior is also used in a self-labeling process to handle low-data regimes, in order to improve the quality of the pseudo-label selection. Results Experiments carried out on CT scans from the public TCIA pancreas segmentation dataset reveal that the resulting STIPPLE model can significantly increase performances compared to the FCN baseline, especially with few training images. We also show that STIPPLE outperforms state-of-the-art semi-supervised segmentation methods by leveraging the spatial prior information. Conclusions STIPPLE provides a segmentation method effective with few labeled examples, which is crucial in the medical domain. It offers an intuitive way to incorporate absolute position information by mimicking expert annotators.