Notice ID: 75N95022Q00054
“Statement of Need and Purpose:
The MRI Section of the NIA/IRP (National Institute on Aging/Intramural Research Program) specializes in studies of tissue response to aging, and age-related pathology. As part of our program in brain mapping in particular, NIA has a need to work with emerging methods in harmonic analysis, machine learning, and data fusion. These will become central elements in our work on non-invasive diagnosis of brain tissue pathology and understanding microstructural changes that occur with aging.”
“Background Information and Objective:
One of the major open questions in aging research is how the brain and brain stem change with age, and what differentiates between healthy and non-healthy aging. This incorporates the development of age-related pathology and disease, including Alzheimer’s disease. The MRI Section has made major advances over the past several years using data stabilization methods. However, new, even more specialized approaches are being developed in the applied mathematics area for signal analysis. These highly mathematical methods are in the realm of novel neural network architectures and implementations and what may be called “data un-compression”, along with data fusion. We wish to apply these emerging methods to our brain MRI work at the NIA IRP after testing on simulated data.”
“Project Requirements/Salient Characteristics:
- Develop mathematical models of how neural network architecture and hyperparameter settings contribute to the efficiency of input-layer-regularized neural networks.
- Based on (1), develop self-regularizing networks for parameter estimation in MR relaxometry and MRI data fusion.
- Construct graph-matching schemes for heterogeneous data fusion in Magnetic Resonance Imaging and other NMR applications, taking advantage of Laplacian embeddings as efficient feature extractors.
- Apply the machine learning methods developed in (1-3) to simulated MR relaxometry and imaging data based on input parameters from published brain, muscle and cartilage studies to compare the accuracy and precision of these methods with the state of the art…”
