Gerard Martí Juan, Ph.D.

Senior AI Researcher | Medical Imaging & Machine Learning

gerardmartijuan@gmail.com GerardMJuan LinkedIn Google Scholar ORCiD

Summary

Research Scientist and AI/ML Engineer with a Ph.D. and over 4 years of postdoctoral experience in developing advanced medical imaging algorithms and ML/AI solutions, focused on analyzing multi-modal brain MRI datasets. Proven expertise in Python and C++, training large models with large amounts of data, multimodal MRI brain sequences, complex image segmentation, registration, and analysis.

Work Experience

Aug 2025 - Present Eden - Senior AI Researcher (Remote, México)

Led AI research initiatives, developing innovative ML and DL solutions to enhance radiology work in a clinical platform impacting thousands of radiologists and millions of patients.
Collaborated cross-functionally with product, engineering, and design teams to successfully integrate AI solutions into production environments.
Mentored team members on AI methodologies while staying current with latest technological advances and industry best practices.

Oct 2022 - Present Universitat Pompeu Fabra - BCN Medtech - Research Scientist (Barcelona, Spain)

Led the end-to-end design, development, and implementation of advanced reconstruction and segmentation algorithms for predictive analytics on T2 low resolution fetal MRI, optimizing data pipelines for large-scale cohorts (over 600 subjects).
Developed and applied advanced analytical techniques, including segmentation, surface reconstruction, unsupervised ML and interpretability methods, for reconstructed fetal MRI (323 subjects), successfully identifying distinct patterns of neurodevelopment across subjects.

Apr 2021 - Oct 2022 Vall d’Hebron Research Institute - Postdoctoral Researcher (Barcelona, Spain)

Engineered and deployed a scalable, cloud-based ML pipeline application for processing structural, diffusion and functional MRI from large multi-center cohorts (697 subjects, 7 centers), generating a re-usable analytical framework for neurological disease research.
Developed, optimized, and applied data-driven computational models using the Reduced Wong-Wang model implemented in C++ to analyze brain activity, identifying significant associations with Multiple Sclerosis biomarkers.
Designed and validated a 3D CNN diagnostic support application in Python for semiautomatic optic nerve lesion detection in T2 fat-saturated MRI sequences, achieving 68.11% balanced accuracy and showing generalization to an independent dataset.

Nov 2016 - Mar 2021 Universitat Pompeu Fabra - BCN Medtech - Ph.D. Researcher (Barcelona, Spain)

Utilized statistical methodologies for big data analysis combining neuroimaging (MRI, hippocampal surface) and genetics on 1448 subjects, generating novel insights into gene-environment interactions relevant to Alzheimer’s disease risk.
Developed novel deep generative models (recurrent variational autoencoder, RVAE) in Python using PyTorch for integrating multimodal time-series biological data in Alzheimer’s Disease, enhancing predictive accuracy and modeling disease trajectories.

Education

Apr 2021 Universitat Pompeu Fabra - Ph.D. in Information and Communication Technologies (Barcelona, Spain)

Thesis: “Data-driven methods to characterize heterogeneity in Alzheimer’s disease using cross-sectional and longitudinal data”
Grade: Cum Laude

Oct 2016 Universitat Autònoma de Barcelona - MSc in Computer Vision (Barcelona, Spain)

Sept 2015 Universitat Politècnica de Catalunya - Bachelor’s degree in Informatics Engineering (Barcelona, Spain)

Selected Publications

See my Publications page for a complete list.

Martí-Juan, G., Lorenzi, M., Sanroma-Guell, G., et al. (2023). MC-RVAE: Multi-channel recurrent variational autoencoder for multimodal Alzheimer’s disease progression modelling. NeuroImage. DOI

Martí-Juan, G., Sastre-Garriga, J., et al. (2023). Using The Virtual Brain to study the relationship between structural and functional connectivity in patients with multiple sclerosis: a multicenter study. Cerebral Cortex. DOI

Martí-Juan, G., Frias, M. et al. (2022). Detection of lesions in the optic nerve with magnetic resonance imaging using a 3D convolutional neural network. NeuroImage: Clinical. DOI

Martí-Juan, G., Sanroma-Guell, G., Cacciaglia, R., et al. (2020). Nonlinear interaction between APOE ε4 allele load and age in the hippocampal surface of cognitively intact individuals. Human Brain Mapping. DOI

Technical Skills

Data Analysis & Engineering

SQL, Data Wrangling & Preprocessing, Data Integration & Flow Architectures
Big Data Processing & Analysis, Statistical Analysis (Bayesian Statistics)
Analysis of Large-scale & Complex Datasets (Clinical, Imaging, Genetic, Real-world Data)
Data Visualization (Matplotlib, Seaborn, Plotly)

Machine Learning & Deep Learning

AI/ML Frameworks: PyTorch, TensorFlow, Keras
Predictive Analytics, Unsupervised Learning, Supervised Learning
Multimodal Modeling, Interpretability & Explainability (SHAP)
Computer Vision, Fine Tuning

Programming & Development

Python, R, SQL, C++, Java
Git, Docker, Singularity, GitHub Actions
Cloud Computing, High Performance Computing (HPC), Parallel Computing
LaTeX

Languages

English (Fluent), Spanish (Native), Catalan (Native)