AI Automation in Scanning Electron Microscopy
summary
Scanning electron microscopy workflow is automated to predict image detection time using XGBoost regression on a dataset from three samples with measured focus, magnification, stigmatism, resolution and beam intensity; result ranks importance as focus and beam intensity highest, magnification and resolution next, stigmatism least. SEM specs cited (1–20 nm resolution, up to 1,000,000×) and the model reduces manual imaging time; next steps are Python integration with SEM software and longrun testing, later robotics for loading.
AI-written summary · about
chapters (8)
- 0:00Introduction to AI in SEM
- 2:34Challenges with SEM workflow
- 3:38AI automation solution proposed
- 4:21Data collection and model selection
- 5:46Key factors affecting detection time
- 6:54Focus and beam intensity automation
- 7:15Future integration and testing plans
- 7:59Q&A session
related talks
39mSustainably Automate Manual Cell-Staining with Innovative Open-Sour...
14mAI-Driven Precision Manufacturing
14mAccessibility Scout: Personalized Accessibility Scans of Built Environments
27mAI-ML and space exploration at NASA
21mFighting Cancer with AI
44mNext-Gen Therapies
44mSustainable Machine Learning by Carl Zeiss AG
16mEye-Trace: Segmentation of Volumetric Microscopy Images with Eyegaze