Communications of the ACM
Scientists Develop Computational Approach to Reduce Noise in X-Ray Data
A graphic depiction of the machine learning model, showing the series of XPCS images (top), which are fed into the machine learning model, yielding the denoised data (bottom) that are used for further analysis
Credit: Brookhaven National Laboratory
Scientists at the U.S. Department of Energy's Brookhaven National Laboratory (BNL) have helped to de-noise synchrotron x-ray experiments computationally.
The researchers devised new methods and models using machine learning (ML) for application to x-ray photon correlation spectroscopy experiments.
The ML model's input data was represented mathematically by a two-time intensity-intensity correlation function, from which the researchers had to ascertain how the model would process the data.
BNL's Anthony DeGennaro said, "Our models can extract meaningful data from images that contain a high level of noise, which would otherwise require a lot of tedious work for researchers to process. We think that they will be able to serve as plug-ins for autonomous experiments, such as by stopping measurements when enough data have been collected or by acting as input for other experimental models."
From Brookhaven National Laboratory
View Full Article
No entries found