UC Riverside Library

This presentation introduces NuclearGuard AI, an interactive nuclear safety dashboard that applies deep learning and anomaly detection to public gamma-ray spectral data. The prototype visualizes radiation spectra, anomaly scores, reconstruction errors, and spatial risk patterns to help identify suspicious radiological signals. Using Streamlit, Plotly, and machine learning models such as Isolation Forest and deep autoencoders, the project demonstrates how AI can support radiation monitoring, interpretable nuclear safety analysis, and operator decision-making.