EES Departmental Seminar: The low-pressure metamorphic gradient along Nova Scotia’s South Shore, revisited - Dalhousie University

Artificial intelligence is increasingly being incorporated into radiology workflows, but both AI systems and human readers can be uncertain, inconsistent, or confidently wrong. Our work explores methods for quantifying and visualizing uncertainty in medical imaging AI, with the goal of making model outputs more interpretable and clinically useful. We are interested in how uncertainty-aware algorithms can help radiologists navigate difficult cases and improve consistency in settings with substantial inter-reader variability. This talk will discuss our efforts to better understand and ultimately improve the interaction between two imperfect decision-makers: the radiologist and the AI. David R. Holmes III, Ph. D. is a Professor of Biomedical Engineering at Mayo Clinic. He studies the methods by which healthcare researchers analyze and interpret large data sets. Dr. Holmes conducts his research from the perspective that data drives decisions.

Expanding understanding of flying rivers and discussing how this knowledge can contribute to water security, decision-making and the valuation of Amazon ecosystem services was the focus of the Internal Seminar “Flying Rivers, Water Security and Economic Valuation of Amazon Ecosystem Services,” held on October 7 at the headquarters of the Permanent Secretariat of the Amazon Cooperation Treaty Organization (PS/ACTO) in Brasília. The event brought together experts in climate, hydrology, ecology and economics, representatives of ACTO Member Countries and partner institutions to discuss the state of knowledge on Amazonian moisture flows, the territories that depend on them and the economic value of the hydroclimatic services provided by the forest. Monitoring, governance, financing and opportunities for regional cooperation were also central to the discussions.

Lawrence Livermore National Laboratory researcher Isaiah Santistevan led the development of three-dimensional simulations exploring how a nuclear device could deflect or disrupt an asteroid threatening Earth. Credit: Blaise Douros/LLNL Lawrence Livermore National Laboratory (LLNL) researchers are developing three-dimensional simulations to examine how a nuclear device could be used to deflect or potentially disrupt an asteroid threatening Earth.