David José Valverde and Jason Moraitis presented their work at the 24th Annual Conference of the International Association for Mathematical Geosciences (IAMG 2026), held in Montreal (Canada) from 23 to 28 August 2026. Jason Moraitis was awarded the Best Student Oral Presentation.
David Valverde presented Foundation-model-assisted mineral grain segmentation on underground mine faces from RGB imagery, developed as part of the European Project MINOTAUR – Miniaturized Robotic Systems for Autonomous In-Situ Exploration of Critical Raw Materials in Deep Land Deposits. The work was co-authored by Gustavo Paneiro, Helga Jordão, Leonardo Azevedo, and Maria João Pereira. It explores the use of artificial intelligence in the identification and segmentation of mineral grains in images of subsurface exploration.
Jason Moraitis presented Propagating Geological Boundary Uncertainty from Bayesian Neural Networks into Conditional Grade Simulations Using Information-Theoretic Measures, co-authored by Helga Jordão as part of the exploration project Advancing Mineral Resource Modeling with Digital Twin Technology. The work was developed in the scope of Jason’s master's thesis during his Erasmus+ internship at Instituto Superior Técnico.
Técnico’s representation at the conference included Maria João Pereira, president of the Department of Energy and Mineral Resources (DER) of Instituto Superior Técnico. The participation in IAMG 2026 reinforced Técnico’s and CERENA’s presence in one of the main international encounters devoted to mathematical geoscience, artificial intelligence, and modeling of mineral resources.