arXiv subjects

A. Ribeiro

Publications and source records attributed to A. Ribeiro.

5 recordsLinked to original sources

arXiv2025

A weakly-supervised deep learning model for fast localisation and delineation of the skeleton, internal organs, and spinal canal on Whole-Body Diffusion-Weighted MRI (WB-DWI)

Background: Apparent Diffusion Coefficient (ADC) values and Total Diffusion Volume (TDV) from Whole-body diffusion-weighted MRI (WB-DWI) are recognized cancer imaging biomarkers. However, manual disease delineation for ADC and TDV measurements is unfeasible in clinical practice, demanding automation. As a first step, we propose an algorithm to generate fast and reproducible probability maps of the skeleton, adjacent internal organs (liver, spleen, urinary bladder, and kidneys), and spinal canal. Methods: We developed an automated deep-learning pipeline based on a 3D patch-based Residual U-Net architecture that localises and delineates these anatomical structures on WB-DWI. The algorithm was trained using "soft labels" (non-binary segmentations) derived from a computationally intensive atlas-based approach. For training and validation, we employed a multi-centre WB-DWI dataset comprising 532 scans from patients with Advanced Prostate Cancer (APC) or Multiple Myeloma (MM), with testing on 45 patients. Results: Our weakly-supervised deep learning model achieved an average dice score of 0.67 for whole skeletal delineation, 0.76 when excluding ribcage, 0.83 for internal organs, and 0.86 for spinal canal, with average surface distances below 3mm. Relative median ADC differences between automated and manual full-body delineations were below 10%. The model was 12x faster than the atlas-based registration algorithm (25 sec vs. 5 min). Two experienced radiologists rated the model's outputs as either "good" or "excellent" on test scans, with inter-reader agreement from fair to substantial (Gwet's AC1 = 0.27-0.72). Conclusion: The model offers fast, reproducible probability maps for localising and delineating body regions on WB-DWI, potentially enabling non-invasive imaging biomarker quantification to support disease staging and treatment response assessment.

cs.CV↗

arXiv2014

Approaching the Asymptotic Regime of Rapidly Rotating Convection: Boundary Layers vs Interior Dynamics

Rapidly rotating Rayleigh-Bénard convection is studied by combining results from direct numerical simulations (DNS), laboratory experiments and asymptotic modeling. The asymptotic theory is shown to provide a good description of the bulk dynamics at low, but finite Rossby number. However, large deviations from the asymptotically predicted heat transfer scaling are found, with laboratory experiments and DNS consistently yielding much larger Nusselt numbers than expected. These deviations are traced down to dynamically active Ekman boundary layers, which are shown to play an integral part in controlling heat transfer even for Ekman numbers as small as $10^{-7}$. By adding an analytical parameterization of the Ekman transport to simulations using stress-free boundary conditions, we demonstrate that the heat transfer jumps from values broadly compatible with the asymptotic theory to states of strongly increased heat transfer, in good quantitative agreement with no-slip DNS and compatible with the experimental data. Finally, similarly to non-rotating convection, we find no single scaling behavior, but instead that multiple well-defined dynamical regimes exist in rapidly-rotating convection systems.

physics.flu-dyn↗

arXiv2011

Accelerated Dual Descent for Network Optimization

Dual descent methods are commonly used to solve network optimization problems because their implementation can be distributed through the network. However, their convergence rates are typically very slow. This paper introduces a family of dual descent algorithms that use approximate Newton directions to accelerate the convergence rate of conventional dual descent. These approximate directions can be computed using local information exchanges thereby retaining the benefits of distributed implementations. The approximate Newton directions are obtained through matrix splitting techniques and sparse Taylor approximations of the inverse Hessian.We show that, similarly to conventional Newton methods, the proposed algorithm exhibits superlinear convergence within a neighborhood of the optimal value. Numerical analysis corroborates that convergence times are between one to two orders of magnitude faster than existing distributed optimization methods. A connection with recent developments that use consensus iterations to compute approximate Newton directions is also presented.

math.OC↗

arXiv2009

Generation of axisymmetric modes in cylindrical kinematic mean-field dynamos of VKS type

In an attempt to understand why the dominating magnetic field observed in the von-Kármán-Sodium (VKS) dynamo experiment is axisymmetric, we investigate in the present paper the ability of mean field models to generate axisymmetric eigenmodes in cylindrical geometries. An $α$-effect is added to the induction equation and we identify reasonable and necessary properties of the $α$ distribution so that axisymmetric eigenmodes are generated. The parametric study is done with two different simulation codes. We find that simple distributions of $α$-effect, either concentrated in the disk neighbourhood or occupying the bulk of the flow, require unrealistically large values of the parameter $α$ to explain the VKS observations.

physics.flu-dyn↗

arXiv2008

Impact of impellers on the axisymmetric magnetic mode in the VKS2 dynamo experiment

In the VKS2 (von Kármán Sodium 2) successful dynamo experiment of September 2006, the magnetic field that was observed showed a strong axisymmetric component, implying that non axisymmetric components of the flow field were acting. By modeling the induction effect of the spiraling flow between the blades of the impellers in a kinematic dynamo code, we find that the axisymmetric magnetic mode is excited and becomes dominant in the vicinity of the dynamo threshold. % The control parameters are the magnetic Reynolds number of the mean flow, the coefficient measuring the induction effect, $α$, and the type of boundary conditions (vacuum for steel impellers and normal field for soft iron impellers). We show that using realistic values of $α$, the observed critical magnetic Reynolds number, $Rm^c \approx 32$, can be reached easily with ferromagnetic boundary conditions. We conjecture that the dynamo action achieved in this experiment may not be related to the turbulence in the bulk of the flow, but rather to the alpha effect induced by the impellers.

physics.flu-dyn↗