My work investigates the intersection of atmospheric remote sensing, machine learning, and geospatial data engineering — so that satellite observations translate into actionable environmental and public-health insight.
NASA · 2026
How often should ML models be retrained to stay accurate? Empirical analysis using NASA GEOS-CF satellite data and local monitoring observations.
Read the paperFUTURE-AI: international consensus guideline for trustworthy and deployable artificial intelligence in healthcare
Establishes evaluation frameworks for trustworthy AI deployment in clinical settings.
medigan: a Python library of pretrained generative models for medical image synthesis
Open-source Python library providing plug-and-play access to pretrained generative models for synthetic medical image generation.
Risk communication of ambient air pollution in the WHO European Region: review of air quality indexes and lessons learned
Comparative review of national AQI frameworks across the WHO European Region, with policy recommendations for health risk communication.
Authored Publications
Loading publications…
No publications match the selected filters.
Talks & Presentations
Loading talks…
Research Sponsored By
Supported by to NASA for artificial intelligence modeling to study atmospheric composition at global scale, and make data products from NASA Earth-observing satellites accessible to the public through open-source tools.
Collaborating with the WHO European Region and NIH-funded initiatives to translate environmental data into public-health risk frameworks.
Supported by the Fulbright Program, Erasmus+, the European Commission, US departement of State, and the Moroccan Ministry of Higher Education for cross-border research exchange.
