|
|
|
Data
|
A system to share and explore human brain mapping data
|
In this project, we will design, prototype, and evaluate a system that enables the exploration, analysis, and dissemination of structural magnetic resonance imaging (MRI) data.
|
|
|
|
|
|
|
|
|
|
|
|
|
Recent advances in imaging technologies have provided human brain mapping researchers with a wide repertoire of experimental approaches to investigate brain structure and function at a variety of spatial scales. Over the entire neuroscience community, a vast amount of human brain mapping data is being gathered. However, lack of a suitable information infrastructure keeps the value and scientific impact of this data from reaching its full potential, as the data is generally used exclusively by the laboratory of origin. Our long-term objective is the construction of a worldwide information management system which will increase the creativity and productivity of neuroscience investigators, as they use shared human brain mapping data to generate and test ideas far beyond those pursued by the data's originators. In this project, we will design, prototype, and evaluate a system that enables the exploration, analysis, and dissemination of structural magnetic resonance imaging (MRI) data. The key goals of this project are:
<p>
1. Development of a Digital Library. We will develop and deploy an operational digital library that will enable users to contribute and retrieve data, as well as explore and analyze its associated metadata. Initial data will be contributed by the International Consortium on Brain Mapping (ICBM). Novel security and privacy technology will be employed to provide maximal image availability subject to the constraints placed on the data by national law, institutional regulations, laboratory/PI preferences, and subject consent forms.
<p>
2. Development of a Data Warehouse Model and Advanced Query Capabilities. We will extend existing data models and query languages to design a human brain data warehouse, based on carefully enriched digital library content. The extensions will allow users to manipulate and compare segmented images, explore spatial relationships among regions, and reason about abstract attributes of structural regions. We will also design an associated content-based retrieval (CBR) system enabling query-by-example functionality over warehouse data in which users submit an example to obtain a set of similar images.
<p>
3. Dynamic Atlas Generation: We will develop techniques that enable users to dynamically aggregate warehouse data into a probabilistic brain atlas defined over subpopulations of interest. Novel query operations will permit users to visualize atlases, access their basic features, reason spatially about the data, and perform statistical comparisons. Query optimization techniques will ensure rapid interaction with atlases and with the warehouse in general.
|
|
|
|
|