NeuroAI • Biomedical Data • Machine Learning
Building technically rigorous AI for neuroscience and health data.
I'm Ghazaleh Ranjbaran, a machine learning specialist focused on EEG, fMRI, multimodal biomedical modeling, and reproducible research systems. I build practical ML workflows that turn complex data into clearer scientific and clinical insight.
My work sits between research and systems: careful model design, signal-aware analysis, and infrastructure that makes advanced work easier to trust and use.
Focus
NeuroAI and Computational Neuroscience
Deep learning and signal-driven workflows for EEG, fMRI, hyperscanning, and multi-brain decoding.
Machine Learning for Biomedical Data
Multimodal pipelines that connect physiological, imaging, and behavioral data to robust model development.
Reproducible Research Systems
Compute-aware workflows, automation, and data handling practices that make advanced analysis usable at scale.
Selected Work
ADHD Modeling with Brain Imaging and Behavioral Data
Predictive modeling for ADHD diagnosis with attention to sex-based disparities and more equitable clinical insight.
Self-Supervised Learning for Neural and Visual Data
Representation learning workflows spanning SimSiam, multimodal pretraining, and low-label research settings.
Explainable and Reproducible ML Pipelines
Technical systems that support interpretability, signal processing, and reliable experimentation across complex datasets.
Current Direction
My research interests center on self-supervised learning, multimodal neural data, responsible AI, and research systems that stay scientifically rigorous while remaining practical to use.