Intelligent Acquisition and Reconstruction for Hyper-Spectral Tomography Systems

Overview

Development stage

In development or acquisitionSource: a) Pre-deployment – The use case is in a development or acquisition status.

High-impact designation

Not high-impactSource: c) Not high-impact

Impact justification

Does not meet definition

Start date

Not reported

Withheld from public reporting

NoSource: a) No

Mission

Topic area

Science

Operational functions

No function tag matched Derived by keyword rules; see methodology

Problem the AI is intended to solve

We will develop artificial intelligence (AI) and machine learning (ML) algorithms to enable dramatic improvements in the throughput and performance of hyperspectral (i.e., multiple energies) computed tomography (HSCT) beamlines at DOE BES Scientific

Expected benefits

We will develop artificial intelligence (AI) and machine learning (ML) algorithms to enable dramatic improvements in the throughput and performance of hyperspectral (i.e., multiple energies) computed tomography (HSCT) beamlines at DOE BES Scientific User Facilities (SUFs). We will demonstrate the utility of our algorithms by carefully designing experiments for energy materials at HSCT beamlines available at the Spallation Neutron Source (SNS) and the National Synchrotron Light Source II (NSLS-II). We will also develop AI driven data acquisition algorithms that will optimize the scanning strategy on-the-fly, in order to obtain the fewest yet most informative set of measurements (i.e. reducing beam time and/or number of projections in a data set). Our team will provide ML based reconstruction algorithms that can produce high quality reconstructions from incomplete, sparse and low signal-to-noise ratio datasets enabling real-time feedback and ensuring best possible reconstruction on completion of the experiment. Finally, our efforts will be available to the user community at both facilities via a general user interface.

System outputs

We will develop artificial intelligence (AI) and machine learning (ML) algorithms to enable dramatic improvements in the throughput and performance of hyperspectral (i.e., multiple energies) computed tomography (HSCT) beamlines at DOE BES Scientific User Facilities (SUFs). We will demonstrate the utility of our algorithms by carefully designing experiments for energy materials at HSCT beamlines available at the Spallation Neutron Source (SNS) and the National Synchrotron Light Source II (NSLS-II). We will also develop AI driven data acquisition algorithms that will optimize the scanning strategy on-the-fly, in order to obtain the fewest yet most informative set of measurements (i.e. reducing beam time and/or number of projections in a data set). Our team will provide ML based reconstruction algorithms that can produce high quality reconstructions from incomplete, sparse and low signal-to-noise ratio datasets enabling real-time feedback and ensuring best possible reconstruction on co

Technology

AI classification

Classical / predictive MLSource: Classical/Predictive Machine Learning: Models trained on data to make predictions or classifications based on identified patterns or relatio…

System name(s)

Not reported

Custom-developed code

Not reported

Public source code

Not reported

Sourcing

How it was built

Not reported

Vendor (as reported)

Not reported

Vendors (standardized)

None on the standard list

A vendor is the supplier named by the agency. It does not identify the underlying model or AI technology.

Data and privacy

Involves PII

Not reported

Privacy Impact Assessment

Not reported

Authorization to Operate

Not reported

Demographic features

Not reported

Training and evaluation data

Not reported

Federal Data Catalog

Not reported

Governance

0 of 8 minimum-practice questions answered. A blank answer means the agency reported nothing; it does not mean the practice is absent.

Potential impacts and how they were identified

Not reported