24-009924 SOW.docx

STATEMENT OF WORK (SOW)

(SERVICES)

GENERAL INFORMATIONTitle of Project:

Development of Multi-modality Multispectral Laplacian Eigenmap Embedding for Super-resolution in MRI Statement of Need and Purpose:

The purpose of this acquisition is to purchase service to develop a graph-based model for multispectral MRI datasets, based on current state-of-the art on related structures for RGB images.

.Background Information and Objective:

The magnetic resonance imaging (MRI) section of the National Institute on Aging (NIA) Intramural Research Program (IRP) specializes in studies of tissue response to aging, and age-related pathology. As part of this program in brain mapping in particular, NIA has a need to work with intrinsically lengthy and relatively low-resolution mapping modalities, including diffusion tensor imaging (DTI). In addition to this, depending on imaging parameters, T1- and T2-weighted images may also require lengthy imaging times and hence may exhibit low resolution. There is an ongoing need to develop methods for increased resolution without corresponding increases in imaging time. These will become central elements in our work on non-invasive diagnosis of brain tissue pathology and understanding microstructural changes that occur with aging.

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 characterization 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, these do not directly address the problem of the tradeoff between imaging resolution and data acquisition time.

Initial progress has been made in this area using the Laplacian eigenmap dimensionality reduction technique, in which features are extracted from low- and high-resolution imaging modalities, matched in an embedding space, and used to recover a high-resolution version of the low-resolution input image. However, this imposes what is in effect an artificial limit on the algorithm, since multispectral (i.e. multi-weighted) image sets are often obtained routinely, and in some instances without additional imaging time (as for a stack of T2-weighted images). Thus, there is a need to extend the current methods to make full use of multispectral images for super-resolution of MRI images. This would provide a means for obtaining improved image quality for our studies of aging and age-related pathology.

Period of Performance:

8/15/2024 – 8/24/2025

SCOPE OF WORKGeneral Requirements:

Independently and not as an agent of the Government, the Contractor shall furnish all the necessary services, qualified personnel, material, equipment, and facilities, not otherwise provided by the Government as needed to perform the Statement of Work below:

Specific Requirements:

The contractor shall provide detailed instructions on using this code so that National Institute on Aging (NIA) researchers can reproduce the work and apply the analytical techniques developed to new data. These instructions will include both annotations within the code and separate documents providing step-by-step instructions.

The contractor shall provide the results of applying the analytical techniques developed to simulated data and to on-line MRI data repositories.

The timely submission of deliverables is essential to successfully completing this requirement. Deliverables are to be submitted in electronic copy.

1) Develop a graph-based model for multispectral MRI datasets, based on our current development of related structures for RGB images, and our work on uni-spectral MRI.

2) Based on (1), perform Laplacian eigenmap embedding.

3) Construct graph-matching schemes for matching embedded images of different resolution, taking advantage of Laplacian embeddings as efficient feature extractors.

4) Develop and apply an appropriate pre-image scheme to derive a final high-resolution version of the low-resolution input image.

5) Implement performance metrics for the procedure through use of ground-truth high-resolution images.

LEVEL OF EFFORT:

Mathematician contractor, 250 hours.

GOVERNMENT RESPONSIBILITIES

The government shall provide post-baccalaureate student support towards development and application of these methodologies. The government shall also provide on-line data repository MRI images for development and application. The project officer will evaluate code produced by the contractor and provide feedback at each of the five stages described above.

DELIVERY OR DELIVERABLES

The contractor shall deliver code (e.g. Python and MATLAB scripts) generated in steps

(1) through (5) above.

The timely submission of deliverables is essential to successfully completing this requirement. Deliverables are to be submitted in electronic copy.

REPORTING REQUIREMENTS

The contractor is required to provide status updates during the period of performance. This shall cover all work completed during the specified period and shall present the work to be accomplished during the subsequent period. The status updates shall also identify any problems that arose and a statement explaining how the problem was resolved. The status updates shall also identify any problems that have arisen but have not been completely resolved and provide an explanation. No formal written report is required.

OTHER CONSIDERATIONSKey Personnel:

Mathematics PhD with experience in advanced machine learning, dimensionality reduction, graph theory, Laplacian eigenmap theory and application, and applications of these to image processing problems.

Data Rights:

The NIA shall have unlimited rights under this contract including in reports, recommendations, briefings, work plans and all other deliverables in accordance with FAR 52.227-14 Rights in Data-General ALT IV. This includes the deliverables provided under the basic contract and any optional task deliverables exercised by the contracting officer. In addition, it includes any additional deliverables required by contract change. The definition of “unlimited rights” is contained in Federal Acquisition Regulation (FAR) 27.401, “Definitions.”

Publications and Publicity:

In collaboration with NIA investigators, the contractor may write and publish articles of data produced as a result of the requirements of the SOW.

Confidentiality of Information:

This is not applicable because only simulated data and on-line databases will be analyzed in this project.

Version 11-30-2016-1 FAR Part 11—Describing Agency Needs