PhD Dissertation Defense

Data-driven group analysis of complex-valued fMRI data

Pedro A. Rodriguez

11:00am Tuesday, 16 April 2013, ITE 346, UMBC

Analysis of functional magnetic resonance imaging (fMRI) data in its native, complex form has been shown to increase the sensitivity of the analysis both for data-driven techniques such as independent component analysis (ICA) and for model-driven techniques. The promise of an increase in sensitivity and specificity in clinical studies provides a powerful motivation for utilizing both the phase and magnitude data; however, the unknown and noisy nature of the phase poses a challenge for successful study of the fMRI data. In addition, complex-valued analysis algorithms, such as ICA, suffer from an inherent phase ambiguity, which introduces additional difficulty for group analysis and visualization of the results. We present solutions for these issues, which have been among the main reasons phase information has been traditionally discarded, and show their effectiveness when used as part of a complex-valued group ICA algorithm application. The developed methods become key components of a framework that allows the development of new fully complex data-driven and semi-blind methods to process, analyze, and visualize fMRI data.

In this dissertation, we first introduce the methods developed as part of the fully complex framework for ICA of fMRI data. We introduce a physiologically motivated de-noising method that uses phase quality maps to successfully identify and eliminate noisy voxels—3D pixels—in the fMRI complex images so they can be used in individual and group studies. We also introduce a phase correction scheme that can be either applied sub-sequent to ICA of fMRI data or can be incorporated into the ICA algorithm in the form of prior information to eliminate the need for further processing for phase correction. Finally, we present two visualization methods that are used to augment the sensitivity and specificity in the detection of activated voxels. We show the benefits of using the developed methods on actual complex-valued fMRI data.

In the remainder of the dissertation, we focus on developing constrained ICA (C-ICA) algorithms for complex-valued fMRI data. C-ICA uses prior information, hence providing a balance between model-based and data-driven approaches such as ICA to improve the source estimation performance and robustness to noise. C-ICA algorithms have been used to improve the estimation performance in real-valued fMRI data, but—to our knowledge—have not been applied to complex-valued fMRI data. We develop the first C-ICA algorithm that uses complex-valued references to constrain either the sources or the mixing coefficients. The designed algorithm is not restricted to having a unitary demixing matrix, which is a major assumption in existing C-ICA algorithms. We show, on both simulated and actual fMRI data, how the performance of ICA improves by using prior information about the fMRI paradigm.

Committee: Dr. Adali (Chair), Dr. Morris, Dr. Rutledge, Dr. Laberge, Dr. Phlypo, and Dr. Calhoun

The CSEE Department will hold a "town hall" meeting for undergraduate COmputer Science (CMSC) majors, minors and other interested students in ITE 456 from 12:00 to 2:00 on Thursday, April 18.

This is an opportunity to interact with your department chair, Professor Gary Carter, the CMSC undergraduate program director Professor Marc Olano and other faculty members. During the meeting you will hear about recent developments in the department and CMSC program, and have opportunities to express opinions, raise issues, make suggestions, ask questions and discuss how to make the CMSC program better. There will also be pizza and drinks.

If you plan on attending, please send an email message to Sorry, you need javascript to view this email address. so we can be sure to order enough food. If you have any questions or topics that you would like to raise in advance, send them to Sorry, you need javascript to view this email address. We look forward to a lively and useful event where the communication flows both ways.

PhD Dissertation Defense

Independent Vector Analysis:
Theory, Algorithms, and Applications

Matthew Anderson

1:45pm Wednesday, 17 April 2013, ITE 325B

The field of blind source separation (BSS) is a well studied discipline within the signal processing community due to its applicability to a variety of problems when the data observation model is poorly known or difficult to model. For example, in the study of the human brain with functional magnetic resonance imaging (fMRI), a neuroimaging sensor, BSS algorithms are able to provide medical researchers and practitioners with a decomposition of a three-dimensional ‘movie’ of the brain that is amenable to analysis. BSS algorithms achieve this decomposition with only a few justifiable assumptions; this is contrary to methods based on the general linear model, which require prespecified models of the expected or desired response to achieve analysis of fMRI data.

Most BSS algorithms consider just a single dataset, but it also desirable to have methods that can analyze multiple subjects or data collections in fMRI jointly, so as to provide insights beyond that achieved with individual analysis of single datasets. Several frameworks for using BSS on multiple datasets jointly have been proposed. The subject of this dissertation is the study of one of these frameworks, which has been termed independent vector analysis (IVA). IVA is a recent extension of the classical independent component analysis (ICA) model to BSS of multiple datasets and it has been the subject of significant research interest. In this dissertation, we provide a formulation of IVA that accounts for sources which possess properties such as a) following Gaussian or non-Gaussian distributions; b) samples are independently and identically distributed (iid) or are dependent; and c) having either linear or nonlinear dependence of sources between datasets. The proposed IVA formulation utilizes the likelihood to define the objective function. This formulation admits to theoretical analysis. In particular, we provide the identification conditions, i.e., we determine when the sources can be ‘blindly’ recovered by IVA, and give a lower bound on the source separation performance.

Several algorithms exist for achieving IVA. We provide several new approaches to developing IVA algorithms and apply these approaches using a Gaussian distribution source model and a more general Kotz distribution model. The former, in addition to leading to efficient IVA algorithms, serves as the distribution model that directly connects canonical correlation analysis (CCA) and ICA.  

Committee: Dr. Tulay Adali (Chair), Dr. Joel Morris, Dr. Aninyda Roy, Dr. Ronald Phlypo, and Dr. Mike Novey

Orbis Technologies, Inc. is currently seekingseveral highly motivated Software Engineering Interns to work at our corporate office in Annapolis, MD.  This position will provide an excellent opportunity to work closely with other developers and company leadership.

Overview/Responsibilities: This position will gain experience in development, implementation, and test of semantic technology solutions for commercial and government clients. The position will assist in developing semantic applications using commercial off-the-shelf (COTS) and Open Source software to create custom solutions for our clients. Potential Applications include: MapReduce programming and Semantic Text Analytics.

Qualifications:

• Pursuing a degree in Computer Science, Software Engineering,  Electrical or related field
• Graduation date expected between December 2013 and December 2014
• U.S. citizenship and willingness to obtain a DoD security clearance
• Strong interest or experience with programming, preferably in Java
• GPA 3.0 or higher on a 4.0 scale (4.0 or higher on a 5.0 scale)
• Demonstrated leadership experienced gained through extracurricular activities or associations.
• Required traits: Self-driven, highly motivated, strong work ethic, avid-learner, team-player, detail oriented, agile, adaptable, and creative at solving problems.

 Program Highlights:

·       Interns gain practical hands-on experience through working directly with cutting-edge technologies and senior management on real-world business and technical problems.

·       Interns are provided a comprehensive overview of the skills needed to operate in a world-leading company through a multi-disciplined structured learning environment that includes weekly presentations by company leadership.

·       Professional training and skill-building in a variety of corporate positions.

·       Many interns are offered the opportunity to work full-time after successfully completing the program and completion of their academic studies.

Compensation: Hourly rate based upon education and experience. 

Application Instructions:  Interested candidates should send an email with the following:

·       Cover letter

·       Resume

·       Current transcripts (unofficial will suffice)

·       A least one letter of recommendation from previous employer or professor (two letters are preferred)

Orbis Technologies, Inc. is a growing technology company based in Annapolis, MD with offices in Warner Robins, GA, Jacksonville, FL and Orlando, FL.  Our company is focused on providing high end Strategic Technology Services to DoD and Fortune 50 Clients. Orbis provides a variety of professional technology services to assist Clients in building next generation information infrastructures including building analytics and applications directly for our customers. This includes hosting private cloud computing infrastructures for select clients who desire security and additional protection of their data. Orbis is a recognized leader in deploying semantic technology based information platforms, semantic text analytics, cloud based analytics and ontological data modeling.

On-Campus Interviews offer students a chance to conveniently interview for professional positions on campus.  Employers travel to our campus for the specific purpose of interviewing preselected students for currently open positions.  This is a great opportunity to interview with employers specifically interested in UMBC students and offers the advantage of a smaller applicant pool.  Interviews take place during the fall and spring. The logistics (viewing positions, applying, signing up for interview time slots, etc.) are all handled via UMBCworks.

Don't miss out on employers coming to UMBC to recruit YOU this fall! The following is a list of employers coming to conduct on-campus interviews with application deadlines within the next month.

**  Insight Global

Position:  Recriter/Account Manager (9257146)
Resume Submission End:  April 10

** Basys

Position:  Associate Software Engineer
Resume Submission End:  April 10

A full list of On-Campus Interviews scheduled is available on our website calendar and on UMBCworks, please check regularly for any updates or changes that may occur.