Data Scientist
Data Scientist

Biography

Sarah Akers is a senior data scientist in the AI and Data Analytics Division at Pacific Northwest National Laboratory (PNNL). Under the Foundational Data Science Group she leads the Translational AI team that transforms cutting-edge artificial intelligence (AI)/machine learning research into practical solutions relevant to the Laboratory’s mission. With a start date at PNNL in the summer of 2014, she has leveraged her bachelor’s training in mathematics and biology along with her master’s degree in industrial mathematics to grow capabilities in laboratory automation, computer vision, and rare or latent signature/attribute detection. She is a technical lead for work in nuclear nonproliferation, several few-shot segmentation applications in materials science for the Mathematics for Artificial Reasoning in Science (MARS) and Chemical Dynamics Initiative (CDi) laboratory directed research and development projects, and serves as a key player in operationalizing mission relevant capabilities.

Disciplines and Skills

  • Deep Learning
  • Image Analysis
  • Machine Learning
  • Python
  • AI Enhanced Automation
  • Statistical Data Analysis
  • Statistical Software

Education

MS in industrial mathematics, Utah State University

BA in mathematics, Carroll College

Patents

  • Stanfill B.A, et al. 2020. “Sensing analytical instrument parameters, specimen characteristics, or both from sparse datasets.” U.S. Patent No. 10,541,109.

Publications

  • Akers, S., et al. 2023. "Maximizing Modalities: Accelerating Quantitative Multimodal Electron Microscopy." Oxford University Press US.
  • Kalinin, S. V., et al. 2023. "Deep Learning for Automated Experimentation in Scanning Transmission Electron Microscopy." arXiv preprint arXiv:2304.02048.
  • Olszta, M., et al. 2023. "Automated Oblique Tilt Series in STEM." Microscopy and Microanalysis 29(Supplement_1): 1874-1874.
  • Oostrom, M., et al. 2023. "Classifying metal‐binding sites with neural networks." Protein Science 32(3): e4591.
  • Oostrom, M., et al. 2023. "Fine-tuning TrailMap: The utility of transfer learning to improve the performance of deep learning in axon segmentation of light-sheet microscopy images." bioRxiv: 2023.2010. 2023.563546.