Skip to content

PhD researcher in computational neuroscience

Ander Fernández Rúa

I build mathematical and computational models of brain networks to understand how they age, how they fail in neurodegenerative disease, and how the body’s metabolism shapes them. Based at the Computational Neuroimaging Lab, BioBizkaia Health Research Institute.

Research

The brain is a network of billions of interacting parts whose collective behaviour we can measure, but still struggle to predict. My PhD treats it the way a physicist treats any complex system: write down a model, fit it to data, and see where it breaks.

Whole-brain network models

Dynamical models running on structural and functional connectomes, to explain how large-scale brain activity emerges from anatomy and how it reorganises with ageing and neurodegeneration.

Brain and metabolism

Coupling physiological models of glucose and insulin regulation to brain network dynamics, to ask how insulin resistance and metabolic state shape neural synchronisation and cognition.

Biological data meets neuroimaging

Integrating gene-expression maps and clinical data with neuroimaging, and using machine learning to find markers of brain health and disease.

About

I keep asking the same question of very different systems: how do many simple parts add up to something that behaves as a whole? I first asked it of a hydrogen plasma, training a classifier to tell its states apart from sensor signals alone. That became my first paper. Then of voters, with Ising models and Markov chains fitted to forty years of Spanish elections. Then of wind turbines wearing out at Ikerlan.

The brain is where that question gets hardest and where the answer matters most. It is noisy, multiscale and tied to the rest of the body, and getting its models right has consequences for how we understand ageing, dementia and metabolic disease. That is what I want to spend my career on.

I trained as both a physicist and an electronic engineer, so I am as comfortable deriving a model as writing the pipeline that tests it against data. When I am not doing either, I am probably reading science fiction.

Experience & education

  1. 2026 – present

    PhD researcher, Computational Neuroimaging Lab

    BioBizkaia Health Research Institute, Cruces University Hospital · PhD in Biomedical Research, UPV/EHU

    Supervised by Jesús M. Cortés and Ivan Carcamo, both Ikerbasque researchers.Mathematical and computational modelling of brain dynamics and connectivity, applied to brain ageing, neurodegenerative disease and the interplay between brain and metabolism.

  2. 2025 – 2026

    Data scientist, AI & Data department

    Ikerlan Technology Research Centre

    Worked on an industrial prognostics and health management (PHM) project for wind turbines, predicting mechanical wear from sensor time series with machine learning. I designed and built an automated pipeline that acquires, processes and makes real-time predictions on experimental wear data, and an interactive tool that lets engineers use those predictions for predictive maintenance.

  3. 2025

    MSc in Big Data & Visual Analytics

    UNIR · Specialisation in artificial intelligence and automated reasoning

    Processing and analysing large datasets, machine learning, and visual analytics: turning complex data into visualisations people can explore and reason with.

  4. 2019 – 2024

    Double degree in Physics and Electronic Engineering

    University of the Basque Country (UPV/EHU)

    • Electronic engineering thesis: machine-learning classification of plasma states in an ECR ion source from sensor signals, published as a first-author paper at ECRIS 2024.
    • Physics thesis: statistical-physics models (Ising-type agents, Markov chains) of Spanish voting behaviour.
  5. 2023

    Research internship

    UPV/EHU, group of Dr. Hegoi Manzano

    Code to compute topological rigidity in disordered materials.

Skills

Modelling

  • Dynamical systems and differential-equation models
  • Network dynamics and synchronisation
  • Statistical physics and agent-based simulation
  • Markov and Monte Carlo methods

Data & learning

  • Time-series and signal analysis
  • Machine learning: random forests, clustering, PCA
  • Neural networks, including recurrent (LSTM)
  • Interactive data visualisation

Programming

  • Python (NumPy, SciPy, scikit-learn, TensorFlow/Keras)
  • MATLAB
  • Fortran
  • Java
  • Git

Languages: Spanish (native), English (C2), Basque (C1), French (B1), German (basic).