Personal archive / selected work
Hi 👋 My name is João Henrique Campagnolo. I am a PhD-trained neuroscientist with a background in biomedical engineering. My work spans stress resilience, behavioral analysis, whole-brain activity mapping, machine learning, and computational research tools.
Current focus
- Computational & systems neuroscience
- Stress resilience and recovery dynamics
- Active inference / Predictive Coding
- ML for biomedical and behavioral data
- BERT / BioBERT for clinical-trial text
Selected work
Research, projects, and teaching.
Stress resilience in larval zebrafish
PhD work on behavioral recovery after acute stress, combining standardized assays, whole-brain activity mapping, multivariate analysis, and Python pipelines. Public-facing material focuses on methods and scope while manuscripts remain in preparation.
View thesis & researchUnsupervised behavior quantification
Master’s thesis work at EPFL on unsupervised quantification of Drosophila behavior from pose and kinematic features, contributing to computational behavior analysis and related DeepFly3D work.
Open repositoryPillProphet
A drug-development prediction project built around structured and unstructured clinical-trial data, combining data collection, NLP-based feature extraction, and predictive modelling.
Open repositoryComputational Neuroscience
Teaching materials and practical content developed for master’s students in computational neuroscience, with a focus on intuition, scientific reasoning, and accessible machine learning.
Open repositorySelected publications
Papers and manuscripts.
DeepFly3D, a deep learning-based approach for 3D limb and appendage tracking in tethered, adult Drosophila
eLife, 2019
Quantifying Stress Resilience as Locomotor Rebound in Larval Zebrafish
Manuscript in preparation
Neural correlates of interindividual variability in rebound from acute stress in the zebrafish
Manuscript in preparation
Background
Scientific training, computational bias.
I am a neuroscientist with a background in Biomedical Engineering and Biophysics. I completed my PhD in Neuroscience at the University of Copenhagen, where I worked in Florence Kermen's lab on the neural correlates of stress resilience in larval zebrafish. My doctoral work combined behavioral assays, whole-brain activity mapping, multivariate statistics, and scientific programming to study how brains recover from acute stress.
Before that, I carried out my master's thesis at EPFL in Pavan Ramdya's lab, where I worked on unsupervised quantification of behavior in Drosophila melanogaster. Across these research settings, I have moved between experiment and computation: behavior quantification, imaging data, machine learning, statistical analysis, and building tools that make scientific questions easier to handle.
I am particularly interested in computational and systems neuroscience, active inference, and machine learning methods applied to biological and biomedical data. I also enjoy independent projects that sit slightly outside my formal research path — from clinical-trial NLP work such as PillProphet to smaller side projects that are somewhere between useful, curious, and unnecessarily committed.
Elsewhere