By Gail Dodge and Tiffany Whitfield

A collaboration between nuclear physicists and computer scientists at Old Dominion University and the Thomas Jefferson National Accelerator Facility (Jefferson Lab) has dramatically increased the information that can be extracted from laboratory experiments, potentially worth millions of dollars a year.

Nikos Chrisochoides, Richard T. Cheng Endowed Professor and Distinguished Researcher in ODU’s Computer Science Department, and his group at the Real-Time Computing Center (CRTC) worked with Jefferson Lab scientist Gagik Gavalian to apply the machine learning (a type of artificial intelligence) to the analysis of nuclear physics data.

At Jefferson Lab, electrons are accelerated to the speed of light, forming a powerful high-energy beam to probe inside the atomic nucleus. The nucleus is made up of protons and neutrons, each of which is further made up of quarks and gluons. When an electron collides with one of these subatomic particles (called an “event”), multiple particles are usually produced, including particles created from the energy of that event. These ejected particles help us understand the substructure of protons and neutrons, where many scientific questions remain unsolved.

The application of machine learning by ODU scientists to identify particle tracks at the Jefferson Lab involved the use of entirely new techniques to solve what was not possible before.

As Lawrence Weinstein, ODU Eminent Scholar and Professor of Physics and former President of the Jefferson Lab Users Organization, wrote: “By improving particle trajectory reconstruction, this project increased the number of complicated nuclear collisions reconstructed by 35 %. This allows us to get 35% more physics from the same data, which would otherwise cost around $5 million a year.”

The challenge for physicists analyzing these electron-nucleus collisions is to identify all the particles created, in particular by measuring their momentum. However, measuring these subatomic particles is difficult. Jefferson Lab uses house-sized “spectrometers” to reconstruct the energies, momentums, and identities of these particles from the tiny traces of energy left behind as they pass through the spectrometers’ detectors.

Unfortunately, in addition to the particle tracks, there is also a lot of noise in the detectors, due to unwanted particles and radiation. Reconstructing a particle trajectory requires identifying all detector “hits” caused by that particle, while ignoring the hits caused by noise. A valid track usually leaves more than 100 hits in the spectrometer. But there may be an equal or greater number of spurious signals among the hundreds of thousands of detection channels.

ODU graduate student Polykarpos Thomadakis and the rest of the CRTC team have developed deep learning models that are used to “de-noise” the very noisy data produced by running the Jefferson’s electron beam Lab and the CLAS12 spectrometer at high rates. After using the denoising method they developed, the particle tracking efficiency increased by up to 80% compared to the conventional algorithm.

“This work suggests that new tracking algorithms with artificial intelligence (AI) open up the possibility of conducting experiments at higher luminosity, collecting larger data samples for physical reactions in a shorter time,” said Chrisochoides.

Gavalian explained the impact: “The expertise provided by the CRTC group has been instrumental in the development of new machine learning models to aid conventional tracking algorithms. The improvements achieved in the efficiency of reconstruction of tracks increase the statistical accuracy of measured observables for existing experimental data, thus the road to conducting experiments with higher luminosity by collecting much more data.”

Chrisochoides and his team at CRTC developed the real-time computing techniques over time at ODU and worked on this project for three years.

“The success of this project highlights the benefits of interdisciplinary research as well as data science techniques, which can be used to enable breakthroughs in other scientific fields,” said Gail Dodge, Dean of the College of Science. of the ODU. “It is not easy for scientists from different fields to learn each other’s language and overcome initial hurdles to create effective collaboration. Dr. Chrisochoides, Dr. Gavalian and the team of CRTC students have made a small financial investment and a lot of hard work. work in an incredible achievement.”

Chrisochoides said: “When compared to the operational cost using software based on traditional data analysis methods and without compromising the accuracy of the results, it doubles the number of physical statistics accumulated per day, allowing experiments to run in half the time.”

What’s next for the CRTC team at ODU? Chrisochoides plans to use quantum computing in particle tracking and brain surgery. Stay tuned.

Related News


Hawkins is one of four United Nations Sigma Liaisons, promoting nursing leadership at UN events and programs. (After)

concussion engineering-espex

As part of their senior synthesis project, the Youth Sports Instrumentation team developed an inexpensive device that could alert trainers to a potential head injury. (After)


The grading system measures the “return on investment in education” for low- and middle-income students. (After)