exist Women 2026 Dorota Bayer
Detecting the unexpected: AI-powered anomaly detection for image-based early warning systems
My project develops AI-based anomaly detection for complex image data. Instead of focusing only on predefined patterns, the system learns what is normal and identifies subtle deviations that may indicate emerging risks. Applied to areas such as satellite-based environmental monitoring, this approach enables earlier and more adaptive responses to critical changes, supporting experts with intelligent early warning systems that make hidden signals visible before they become major problems.
Academic field: Machine Learning and Data Science applied to image analysis and Earth observation
Vision:
My vision is to build a new class of AI systems that help us detect what we do not yet know to look for. Starting with satellite and environmental data, I aim to develop scalable anomaly-detection tools that enable earlier awareness of critical changes. In the long term, this approach could be applied across multiple domains, from medical imaging to industrial monitoring, supporting more proactive, data-driven decision-making in systems where early signals are critical.
My business idea will work because existing monitoring systems focus on known patterns, while this approach enables the detection of unknown and emerging signals in rapidly growing image datasets.
Personal superpower:
Seeing what others overlook — identifying subtle, unexpected signals in complex data and turning them into meaningful insights.