Understanding Multimodal Brain Imaging

After more than a decade of human connectome with a large focus on data collection, now it is the time to sift through this big data and to learn more about neurodegenerative diseases through the use of high performance computing, computer vision, graph analysis, visualization and advanced user interfaces. This is the mission of a new collaboration between academia (UCSF, UCB), national lab (LBNL) and industry (Oblong). Check the video showing the first steps toward our goals [NEWS] [VIDEO]

R data analysis

R programming and environment has turned the "lingua franca" is statistical computing. We have designed applications using this great tool and summarized our endeavour in the presentation "Data analysis using the R project for statistical computing" for a set of exploratory data analyses, issues in dealing with large datasets, high performance computing discussions, plus hints, codes and several links to other tutorials. (More information)

Machine learning

Dealing with huge amounts of data is part of daily life of most of the researchers in LBL. We are back to the activities of LBL CRD Machine Learning Reading Group to discuss strategies to organize and understand our data using algorithms in data mining, exploratory data analysis, visualization, etc. Other interests involve high performance computing and ubiquitous platforms to evaluate feasibility limits for data mining based on current processing power. (More information)

MicroCT analysis

High-resolution images have turned into a common input to data analysis in several scientific domains. This process involves quantification over large data files, containing hundreds of slices of 3D objects, e.g. to measure the porosity of these volumes. Our goal is to design nondestructive techniques to quantify properties in the interior of solid objects, including information on their 3D geometries. This quantification will support modeling of the fluid dynamics into the pore space of the host object.

(More information)


Accelerator models

Numerical simulations of laser-plasma wakefield (particle) accelerators model the acceleration of electrons trapped in plasma oscillations (wakes) left behind when an intense laser pulse propagates through the plasma. This physical phenomenon can be simulated, generating large datasets. We roll up sleeves to search for compact group of electrons under acceleration using automated pattern searching.

(More information)


In January 2009, University of California news reported that high-performance computing and the humanities are connecting at the University of California, San Diego with help from the Department of Energy (DOE) and the National Endowment for the Humanities (NEH). Nersc has provided computing resources and assistance to the project entitled "Visualizing Patterns in Databases of Cultural Images and Video", which may revolutionize the way we contemplate art.

Check some images on flickr. ( More information )

Computer Vision

There are lots of image processing, analysis and pattern recognition projects going on in parallel, mostly with the collaboration of Math Group. Check more at:

  • Computer vision applied to medical images (2010).

  • Check some of our past work in computer vision applied to SAR images.