Artificial Intelligence for Mineral Exploration Laboratory
AIMEX Lab is a research and innovation initiative founded and directed by
Dr. Muhammad Amar Gul, focusing on the integration of artificial intelligence, geochemistry, ore-system science, and mineral exploration.
The lab develops interpretable, data-driven, and reproducible AI workflows to support mineral discovery, metallogenic understanding, and exploration decision-making.
📍 Based in Saudi Arabia (Hybrid) | 🌍 Open to global research & consulting collaborations
📧 amar_geologist@yahoo.com
🔗 LinkedIn · Google Scholar · ResearchGate
AIMEX Lab (Artificial Intelligence for Mineral Exploration) is dedicated to:
- Interpretable AI/ML applied to Earth-science and geochemical data
- Predictive modeling of mineral systems and tectonic environments
- Reproducible, scalable workflows for mineral exploration and research
Operating at the interface of academia and industry, AIMEX Lab addresses complex exploration challenges using data-centric and domain-aware AI methodologies.
AIMEX Lab consolidates both AIMEX-led projects and selected academic and industry research conducted by its director, where aligned with the lab’s AI-driven mineral exploration mission.
Mission
Apply transparent, interpretable, and physics-informed machine-learning techniques to deep-Earth datasets to improve understanding and prediction of mineral deposits.
Vision
Establish reproducible, open, and interpretable AI systems that accelerate global mineral exploration and metallogenic research.
I am a multidisciplinary geoscientist and AI practitioner with a PhD in Geology and Artificial Intelligence / Machine Learning, specializing in:
- Ore-deposit modeling using geochemical and isotopic data
- Large-scale geochemical big-data mining
- Machine-learning and deep-learning applications to trace-element systems
- Geospatial analysis and remote sensing for mineral targeting
I currently work as Project Geologist (AI & Big Data) at China National Geological & Mining Corporation, Saudi Arabia, with combined experience in academic research, national programs, and industry-scale mineral exploration.
https://github.com/Dr-Amar/Pyrite-Gunga-Pb-Zn-Deposit--Machine-Learning
Machine-learning-driven classification of Pb–Zn deposits using pyrite trace-element and isotope geochemistry.
Methods: Random Forest, Gradient Boosting, SVM, MLP
Performance: >94% accuracy (LOGO validation)
Explainability: SHAP, t-SNE, UMAP
https://github.com/Dr-Amar/Sphalerite-Gunga-Pb-Zn-DeepLearning
Deep-learning analysis of sphalerite geochemistry and isotopes for mineralization-zone classification.
Key outcomes: Identification of CD-type mineralization and critical-metal enrichment (Ge, Cd, Ag, Sb)
Deep-learning models applied to sphalerite trace-element and isotopic data to classify mineralization zones in sediment-hosted Pb–Zn systems.
CNN and XGBoost models trained on >5,000 LA-ICP-MS pyrite analyses from 50+ global localities, achieving >95% classification accuracy.
Compilation of ~5,200 analyses from 138 deposits worldwide. RF, GB, MLP, and SVM models used for deposit-type discrimination with AUC > 0.99, supported by SHAP-based interpretation.
ML-based classification of galena from 37 global Pb–Zn deposits using imbalance-aware workflows (SMOTE, RUC). Best Gradient Boosting model achieved 98.19% accuracy.
Python- and Gradio-based tools for real-time deposit-type prediction using mineral-chemistry inputs, designed for explainability-ready decision support.
Landsat-8/9-based alteration mapping across ~600,000 km² using supervised classification and AI-assisted prospectivity modeling.
Spatial mapping and economic assessment of mineral deposits and prospects across Saudi Arabia to support exploration planning, investment prioritization, and sustainable mining strategies aligned with Saudi Vision 2030.
Contribution to the Saudi Vision 2030 NGD program, supporting data modeling, ETL pipelines, QA/QC, ISO/OGC-compliant geospatial standards, and cloud-based, AI-ready geoscience infrastructure.
- Publications in Mathematical Geosciences, Journal of Geochemical Exploration, Geoscience Frontiers, Ore Geology Reviews
- Deep-learning models (CNN, MLP) for tectonic-environment discrimination
-
95% accuracy in deposit-type classification using pyrite, sphalerite, and galena
- Explainable AI (SHAP) applied to large geochemical datasets
- Advanced resampling strategies for class-imbalanced geochemical data
| Domain | Tools & Methods |
|---|---|
| Data Science & AI | Python, scikit-learn, TensorFlow, PyTorch |
| Visualization | Matplotlib, Seaborn, Power BI, Tableau |
| Geochemistry | LA-ICP-MS, EPMA, ioGAS, XRD, XRF |
| Remote Sensing & GIS | ENVI, Google Earth Engine, QGIS, ArcPy |
| Development & Deployment | GitHub, Gradio, Jupyter, VS Code |
-
Big Data Mining on Galena Geochemistry Using Machine Learning Algorithms:
Implications for Metallogenic Discrimination
Gul, M.A., Kanwal, A., Yang, X., Zhang, H.S., Faisal, M. (2026)
Mathematical Geosciences — Accepted
Impact Factor: 3.6 -
Artificial Intelligence–Driven Metallogenic Typing of Pyrite from Global Ore Systems
Gul, M.A., Kanwal, A., Yang, X., Zhang, H.S., Faisal, M. (2026)
Journal of Geochemical Exploration — Accepted -
Ore Genesis and Critical-Metal Enrichment Using Deep Learning
Gul, M.A., et al. (2025)
Journal of Geochemical Exploration
https://doi.org/10.1016/j.gexplo.2025.107771 -
Machine-Learning-Driven Classification of Pb–Zn Ore Deposits Using Pyrite Geochemistry
Gul, M.A., et al. (2025)
Journal of Geochemical Exploration
https://doi.org/10.1016/j.gexplo.2025.107693 -
Big-Data Mining of Galena and Pyrite Geochemistry
Gul, M.A., et al.
Manuscripts under review in Lithos, Gondwana Research, and Mathematical Geosciences
If you use AIMEX Lab workflows, datasets, or code in academic or applied research, please cite the relevant publication and acknowledge:
AIMEX Lab – Artificial Intelligence for Mineral Exploration
Dr. Muhammad Amar Gul
AIMEX Lab welcomes collaboration in:
- Academic research
- Industry consulting and applied AI
- Data benchmarking and AI-tool development
📧 amar_geologist@yahoo.com
🌍 Available for remote and international engagements
© AIMEX Lab — Dr. Muhammad Amar Gul
Artificial Intelligence for Mineral Exploration · Interpretable · Reproducible · Data-Driven
TTransforming geoscience with data, innovation, and purpose