Center for Bio-Image Informatics

Engineering, Biology and Computer Science, working together.

  • Increase font size
  • Default font size
  • Decrease font size
REU 2009-2010

REU 2009-2010

E-mail Print PDF

The Research Experience for Undergraduates (REU) program provides research exposure for undergraduate students and is funded by the National Science Foundation(NSF). Students get involved in the ongoing research projects in the UCSB center for BioImage Informatics by active interaction with the members. The following students are enrolled in the REU program 2009-2010 and a brief description of their project goals is highlighted below. For additional information about REU refer to the following link.


Name: Young Chun Ahn Mentor: Pratim Ghosh Advisor: Dr. B.S. Manjunath Department: Electrical and Computer Engineering, UCSB

Young Chun Ahn





I am Young Chun Ahn, a senior in the Electrical Engineering department at UCSB. Currently, I am working on developing image segmentation by using K-means and mean-shift clustering algorithms. For this project, I am working with my mentor Pratim Gosh. Our goal is to implement image processing algorithms to achieve a better understanding of biological images of cells, tissues, and microtubules. As a long term goal, we hope to apply these novel algorithms to an object recognition and tracking.





Name: Kenny Akametalu

Mentor: Pradeep Koulgi Advisor: Dr. B.S. Manjunath Department: Electrical and Computer Engineering, UCSB


Kenny Akametalu






My research primarily entails looking at the mean shift clustering algorithm. The mean shift algorithm is an algorithm that uses a feature space and search windows in order to locate dense regions (clusters) within that feature space. The algorithm has applications in object tracking, and image segmentation. As of now I am looking at the applications of the algorithm in color image segmentation. I am currently reading papers on the algortihm and looking for ways in which the algorithm can be improved for general applications in color image segmentation.





Grace Leung Mentor: Aruna Jammalamadaka Advisor: Dr. B.S. Manjunath Department: Electrical and Computer Engineering, UCSB

Grace Leung






For my research, I plan to use clustering techniques to track the growth of cells or cellular structures. The clustering methods I will be focusing on is k-means algorithm and mean shift algorithm. Though the implementation of these algorithms, my goal is to find methods to optimize them so that they can produce the same results faster. From there, I will use the optimized methods to detect and track cells as they grow in hopes to detect developmental abnormalities.










Name: Elias Flores

Mentor: Swapna Joshi Advisor: Dr. B.S. Manjunath Department: Electrical and Computer Engineering, UCSB

elias flores



My mentor, Swapna is collaborating with a psychologist Dr. Grafton, here at UCSB. They are interested in identifying differences between brain anatomy of psychopaths and non-psychopaths. Brain image data was collected from New Mexico ail which consisted of teens and adults who volunteered to have their brains scanned by an MRI. A Psychopathy Checklist-Revised (PCL-R) score has been given to each inmate to determine psychopathy. The main goal of the project is to classify psychopaths versus non-psychopaths; however currently we do not have a control set to compare against. Hence we are now interested in developing a regression model hoping to capture a region that changes with respect to the PCL-R score. As a short term goal I will be working on implementing a kernel regression and kernel ridge regression algorithms to help my mentor compare her new algorithm against.






Name: Sevada Abraamyan

Mentors: Emre Sargin. Advisor: Dr. B.S. Manjunath Department: Electrical and Computer Engineering, UCSB


sevada


I am a fourth year undergraduate student studying computer science at the University of California Santa Barbara. During the summer, I have been working with the Bio-Image Informatics Lab at UCSB on developing algorithms to estimate and count the number of moving objects in a video sequence. In particular, I have been counting pedestrians on sidewalks and bicyclists on bike paths. My estimation algorithm divides a video frame into a grid and computes a feature vector composed of twenty low level image features form each cell in the grid. Then using neural networks a function is learned that maps these feature vectors to an estimate of the number of people in each block. I have validated the accuracy of my algorithm in counting pedestrians within the UCSD crowd counting dataset. Following this work I have also developed an algorithm in which tripwires are used to count the number of bicyclists passing on a bike path. At the moment I am working on an algorithm to fit a pictorial model onto the foreground pixels associated with bicyclists on bike paths. I hope to be able to detect and count bicyclists using this model fitting technique. 







Name: Caleb Aguirre Mentors: Vignesh Jagadeesh Advisor: Dr. B.S. Manjunath Department: Electrical and Computer Engineering, UCSB

caleb




I am Caleb Aguirre, a junior in the Computer Engineering program at UCSB. I am a transfer student from Santa Barbara City College. My goal for this internship is to gain a fundamental understanding of object detection in images. To get there, me and my partner (Sevada Abraamyan) hope to successfully implement algorithms to examine an image and properly distinguish between simple objects.







Name: William Moy

Mentor: Nicholas Larusso. Advisor: Dr. Ambuj Singh Department: Computer Science, UCSB

I am a second-year Computer Engineering student at UCSB. I am working with Nick Larusso and Melissa Gross on developing algorithms to automate the analysis of live-cell biological images. Our current focus is on the detection and classification of nuclei in different development stages.


Name: Rotem Raviv Mentors: Panuakdet Suwannatat Advisor: Dr. Tobias Hollerer Department: Computer Science, UCSB

Rotem Raviv




The project would involve all the tasks that are related to representing data with uncertainty on XML: 1) the design of XML schema to support data from biomedical imaging domain 2) the design and development of GUI tools for extending the schema to support other domains (e.g. computer network) 3) the converter units to convert data into XML, and to convert from XML to visual representation This project will play a crucial role in the general framework to visualize data with multiple levels of uncertainty cross domains. The design of XML schema will be a collaboration between the student, the mentor and other team members. The main contribution from the student at this step will be the development of the GUI tool. Future contributions can include developing/implementing techniques for uncertainty visualization, both on normal screens and on a screen array (iWall) when the hardware is ready.





Name: Melissa Gross Mentor: Nicholas Larusso Advisor: Dr. Ambuj Singh Department: Computer Science, UCSB

Melissa




My name is Melissa Gross, and I am a third year Computer Science student in the College of Engineering at UCSB. Research topics within Computer Science that I am interested in include artificial intelligence and computer vision. I am currently working with William Moy and our mentor, Nick Larusso, on a project to automate image data mining. Our data set is comprised of image stacks that give us 3-dimensional information of our subjects, which are of the c. elegans species. Using MATLAB, we hope to be able to locate the nuclei of each cell within each subjectís germ line, after which we will be able to run analysis on the transition of various types of cells within the germ line.









Name: Mary E. Canders Advisors: Dr. Steve Fisher and Dr. Geoff Lewis Department: Department of Molecular, Cellular and Developmental Biology, UCSB


Mary




For my research I will be working with Steve Fisher and Geoff Lewis on a research project whose main goal is to document the remodeling events of neurons and cells at the posterior of the retina following retinal detachment. We will examine the patterns and distribution of cells and their processes in the normal retina and observe how the patterns change following retinal detachment. To accomplish this, I will be using high-resolution electron tomography and confocal microscope imaging of the entire retinal surface. For the tomography project I will trace electron micrographs collected from a series of images in the Z-plane and reconstruct a 3 dimensional image. For the confocal imaging project I will immunohistochemically label pieces of rabbit retina with antibodies to proteins known to be present in the retinal cells. Information from these 2 projects should add to the understanding of the connectivity of retinal cells and how this connectivity changes with retinal detachment.





Name: Alex Wolf

Mentor: Brian Ruttenberg Advisor: Dr. Ambuj Singh Department: Computer Science, UCSB


My name is Alex Wolf, and I am a junior in the Electrical and Computer Engineering department at UCSB. I am currently working with Brian Ruttenberg on developing image processing and pipelining techniques in order to classify and extract information about astrocyte cells more efficiently and effectively. Once we are able to extract the data, the features from the images will be placed into a database and used to classify and relate features to both one another and to their spatial locations in the retina, providing us with valuable information about these cells that we know so little about.