School of Computing & Information Sciences
Gabriel Lizarraga developed a neuroimaging computer application for storing, organizing, and processing data, used by several research centers, including the Miami Beach Mount Sinai Medical Center, 1Florida ADRC, Nicklaus Children’s Hospital, Baptist Hospital of Miami, University of Florida in Gainesville, and the University of Miami. Gabriel became a member of the NSF-funded CATE Center for Advanced Technology and Education) center at FIU since 2008. He participated in the PIRE program, traveling to Barcelona, where he coded for Marenostrum (the most powerful supercomputer in Spain), the IBM Watson Center in New York and the Fluminense Federal University in Rio de Janeiro.
Gabriel received honorable mention in 2001 from the NSF GRFP Fellowship in 2011 and served as a reviewer for the SoutheastCon in Norfolk, VA, in 2016. He finished his bachelors, graduating Magna Cum Laude, in December 2009, and was awarded with the “Outstanding Graduate Award”, finishing his master’s degree in 2010. Early in his PhD studies he received the GAANN Fellowship for two years. Gabriel’s PhD research culminated in the creation of a multi-server web-based application: Neuroimaging Web Services Interface (NWSI), hosted in a secure website at FIU and providing a multitude of services for processing MRI, PET, DTI, and other brain images, with an emphasis in multimodal imaging and data sharing in Alzheimer’s disease and Epilepsy.
Structural and functional brain images are generated as essential elements for medical experts to learn about the different functions of the brain in its normal and pathology states, provide diagnosis and prevention measures, and tailor treatment plans that are subject specific. These images are typically visually inspected by experts in the field and stored as part of a patient’s medical history. Many software packages are available to process medical images, but they are complex and difficult to use for non-computer experts. The software packages are also hardware intensive, requiring specialized servers. Furthermore, the results obtained after processing vary depending on the native operating system as data processed in one system cannot typically be combined with data acquired from another system. As a consequence, this dissertation proposes a novel Neuroimaging Web Services Interface (NWSI) as a series of processing pipelines for a common platform to store, process, visualize and share data with the research community.
The NWSI platform, made up of a series of interconnected servers, is password protected, and is securely accessible through a HIPPA compliant web interface from anywhere in the world. The web-interface driving NWSI is based on Drupal, a popular open source content management system. Drupal provides a user-based platform, in which the core code for the security and design tools are updated and patched frequently to address vulnerabilities, as well as to add new functionalities. New features can also be added to Drupal via modules which can be integrated with Drupal core code, allowing new code to run on Drupal, while maintaining the core software secure and intact. The webserver architecture allows (1) visualization of results by embedding interactive visual analysis of multi-faceted neuroimaging data, and (2) downloading tabulated data results for further processing. Several forms are also available on the interface to capture demographic and clinical data. The processing pipeline starts from a FreeSurfer (FS) reconstruction of T1-weighted MRI images. Subsequently, PET, DTI, and fMRI images can be uploaded and registered to the T1 reference image and viewed directly on the interface using our image browser.
The Webserver captures the uploaded images and performs essential functionalities, but the processing is done on the supporting servers. The computational platform is responsive and scalable; more servers can be added if the workload increases. The current pipeline for PET processing calculates all regional Standardized Uptake Value ratios (SUVRs). The FS and SUVR calculations have been validated using Alzheimer’s Disease Neuroimaging Initiative (ADNI) input images, and results posted at Laboratory of Neuro Imaging (LONI). In an effort to confront the challenges of data comparability across sites, the NWSI system provides access to a calibration process through the centiloid scale, consolidating as a consequence Florbetapir and Florbetaben tracers in amyloid PET images. The interface also offers onsite access to machine learning algorithms for data classification and introduces new heat maps that augment and facilitate expert visual rating of PET images. The system has been piloted using data and expertise from Mount Sinai Medical Center, the 1Florida Alzheimer’s Disease Research Center (ADRC), Baptist Health South Florida, Nicklaus Children’s Hospital, and the University of Miami Center for Cognitive Neuroscience and Aging. All results were obtained using our processing servers in order to maintain data validity, consistency, and minimum processing bias due to mixing outputs from dissimilar servers.