Dani Ushizima

Investigations in nano and micro structures from data coming from national laboratories instruments, and simulations, including collaboration with industry and academia to discover motifs that enable image understanding and decision making At Berkeley Lab, she is a Staff Scientist at the Computational Research Division of LBNL. She is also one of the 2015 DOE Early Career Awarded scientists, a co-PI in Image Analysis/Machine Vision for the Center for Advanced Mathematics for Energy Related Applications (CAMERA).

At the University of California, Berkeley, she is one of the selected BIDS Data Scientists fellows since 2014. Her work focuses on image analysis and pattern recognition applied to diverse scientific domains - images range from biomedical micrographies to geological materials and composites, e.g. micro-tomography of materials with applications to carbon sequestration. She has acted as Principal/Co- Investigator of several projects related to image analysis, machine learning, pattern recognition, content-based image retrieval and high performance computing. Interests include computer vision, quantitative microscopy, and data sciences. [Previous work]

Scientific Image Analysis - hover on these pictures for details:

Synchrotron-based X-ray micro-tomography for analysis of composites with applications to jet engine construction.

Micro-CT of glass beads in biogenic mixture, using microbe S.pasteurii for calcite precipitation in research about efficient carbon sequestration.

Identi cation of palladium faces and platinum core from electron tomography for precise control of catalytic reactions during material design.

Quantitative Structure Activity Relationship (QSAR) models for nanoparticles: quantification of chemical composites.

Segmentation of cervical cells as part of the Pap-smear analysis automation process: this work was awarded 1st place in ISBI'2014 Cervical Cell Recognition Challenge.

Team from LBL/UCSF/Oblong developed a gesture-based interface with network diagrams that show traffic patterns, combined with maps of brain structure.

Using time series of 3D confocal imagery to measure cell movement patterns and speed during mitosis of human mammary epithelial cells (in vivo).

Using high-resolution imagery from Auer's lab to search for sub-cellular structures, such as microtubules within human mammary epithelial cells.

Recent News:

  • 02/2017: Our book is available online - reproducible research in action
  • 02/2017: Research and development in constrained environments (TechWomen in Kenya)
  • 01/2017: Our article on ammonia sensor design reached 1700 views
  • 08/2016: Looking for the faces of scientific images - we got the tool
  • 06/2016: Scientific diplomacy through TechWomen, an Initiative of the U.S. Department of State's Bureau of Educational and Cultural Affairs
  • 02/2016: XData Team is up and inventing - welcome our Visiting Scholars and New Students

  • 08/2015: Women in Science and Engineering, Software Carpentry and WSEC to deliver tutorials on fundamentals of Data Science

  • 05/2015: Ushizima Receives DOE Early Career Award 5-Year Project Aimed at Finding Hidden Info in Images from Experiments

  • 05/2015: LBNL and Software Carpentry deliver tutorials on fundamentals of Data Science

  • 02/2015: Black Girls Code - taking programming to the next level - hackathon

  • 08/2014: Ushizima awarded BIDS fellowship to be a BIDS data scientist at UC Berkeley

  • 08/2014: Ushizima's PhD student, Kate Odziomek, is awarded the Americal Chemical Society for Scienfic Excellence

  • 06/2014: LBNL CS Summer student lecture series

  • 05/29/2014: CAMERA Lecture Series at the ALS

  • 05/2014: Brazilian institution recognition of public service

  • 04/2014: Ushizima and team awarded 1st Place in Algorithm Challenge

  • 2014: The "Math Foundry" is real - meet our DOE center, a.k.a. CAMERA

  • 2014: Showing students the importance of work at LBNL Slides [HERE]

  • 2013: Future brain viewers in UX Magazine: exploratory visualization in medicine;

  • 2013: Exploratory analysis of the brain: video of collaboration between UCSF neurologists, LBNL computer scientists and Oblong industry experts;

  • 2013: Brain, precision-medicine and visualization: national laboratory is ready to play a role in the Brain Initiative related projects;

  • SAMSI 2012-2013: Program on Statistical and Computational Methodology for Massive Datasets

  • 2010: This manuscript is #20 among Top 25 Hottest Papers in Digital Signal Processing. More

    Previous work:

    Before joining LBNL in 2007, she was a Computer Science Professor at the Catholic University of Santos, Sao Paulo, Brazil. While professor, she was the Principal Investigator of Computer Vision in Leukemia Diagnosis (Young Researcher Award) and FAP-Books (FAP-IV), co-PI in Agriculture and Ceramics projects with industry (Small Business Incentive Grants), all sponsored by FAPESP science foundation, Sao Paulo, Brazil. Her PhD is from the University of Sao Paulo (USP) in Computational Physics (2004), where she developed a prototype for computer-aided leukemia diagnosis in collaboration with the Clinic Hospital FMRP-USP, and feature selection tools for general purpose data applications. As part of her PhD, she was also as a Visiting Researcher in the Electrical and Computer Engineering Department at UC Santa Barbara (2004).
  • Intelligent Image Analysis