Known as HEAT-ML, the new AI could lay the foundation for software that significantly speeds up the design of future fusion systems. Such software could also enable good decision-making during fusion operations by adjusting the plasma so that potential problems are thwarted before they start.

“This research shows that you can take an existing code and create an AI surrogate that will speed up your ability to get useful answers, and it opens up interesting avenues in terms of control and scenario planning,” said Michael Churchill, co-author of a paper in Fusion Engineering and Design about HEAT-ML and head of digital engineering at PPPL.

Fusion, the reaction that fuels the sun and stars, could provide potentially limitless amounts of electricity on Earth. To harness it, researchers need to overcome key scientific and engineering challenges. One such challenge is handling the intense heat coming from the plasma, which reaches temperatures hotter than the sun’s core when confined using magnetic fields in a fusion vessel known as a tokamak. Speeding up the calculations that predict where this heat will hit and what parts of the tokamak will be safe in the shadows of other parts is key to bringing fusion power to the grid.

“The plasma-facing components of the tokamak might come in contact with the plasma, which is very hot and can melt or damage these elements,” said Doménica Corona Rivera, an associate research physicist at PPPL and first author on the paper on HEAT-ML. “The worst thing that can happen is that you would have to stop operations.”

PPPL amplifies its impact through public-private partnership

HEAT-ML was specifically made to simulate a small part of SPARC: a tokamak currently under construction by CFS. The Massachusetts company hopes to demonstrate net energy gain by 2027, meaning SPARC would generate more energy than it consumes.

Simulating how heat impacts SPARC’s interior is central to this goal and a big computing challenge. To break down the challenge into something manageable, the team focused on a section of SPARC where the most intense plasma heat exhaust intersects with the material wall. This particular part of the tokamak, representing 15 tiles near the bottom of the machine, is the part of the machine’s exhaust system that will be subjected to the most heat.

To create such a simulation, researchers generate what they call shadow masks. Shadow masks are 3D maps of magnetic shadows, which are specific areas on the surfaces of a fusion system’s internal components that are shielded from direct heat. The location of these shadows depends on the shape of the parts inside the tokamak and how they interact with the magnetic field lines that confine the plasma.

Creating simulations to optimize the way fusion systems operate

Originally, an open-source computer program called HEAT, or the Heat flux Engineering Analysis Toolkit, calculated these shadow masks. HEAT was created by CFS Manager Tom Looby during his doctoral work with Matt Reinke, now leader of the SPARC Diagnostic Team, and was first applied on the exhaust system for PPPL’s National Spherical Torus Experiment-Upgrade machine.

HEAT-ML traces magnetic field lines from the surface of a component to see if the line intersects other internal parts of the tokamak. If it does, that region is marked as “shadowed.” However, tracing these lines and finding where they intersect the detailed 3D machine geometry was a significant bottleneck in the process. It could take around 30 minutes for a single simulation and even longer for some complex geometries.

HEAT-ML overcomes this bottleneck, accelerating the calculations to a few milliseconds. It uses a deep neural network: a type of AI that has hidden layers of mathematical operations and parameters that it applies to the data to learn how to do a specific task by looking for patterns. HEAT-ML’s deep neural network was trained using a database of approximately 1,000 SPARC simulations from HEAT to learn how to calculate shadow masks.

HEAT-ML is currently tied to the specific design of SPARC’s exhaust system; it only works for that small part of that particular tokamak and is an optional setting in the HEAT code. However, the research team hopes to expand its capabilities to generalize the calculation of shadow masks for exhaust systems of any shape and size, as well as the rest of the plasma-facing components inside a tokamak.

DOE supported this work under contracts DE-AC02-09CH11466 and DE-AC05-00OR22725, and it also received support from CFS.

ECU PhD student Mr Francesco Bettariga found that a single bout of exercise increased the levels of myokines, a protein produced by muscles which have anti-cancer effects, and which could reduce the proliferation of cancer growth by 20 to 30 per cent.

“Exercise has emerged as a therapeutic intervention in the management of cancer, and a large body of evidence exists that show the safety and effectiveness of exercise as medicine, either during or post cancer treatment,” Mr Bettariga said.

His research with survivors of breast cancer measured myokine levels before, immediately after and 30 minutes post a single bout of either resistance of high intensity interval training and found that both sets of exercise had a resultant increase in myokine levels.

While higher levels of myokines were expected in a healthy population, post a vigorous workout, Mr Bettariga investigated whether breast cancer survivors would see the same results, given the impact that cancer treatments and cancer itself often has on the body.

“The results from the study show that both types of exercise really work to produce these anti-cancer myokines in breast cancer survivors. The results from this study are excellent motivators to add exercise as standard care in the treatment of cancer,” Mr Bettariga said.

He added that the long-term implications of elevated myokine levels should be further investigated, particularly in relation to cancer recurrence.

Further research by Mr Bettariga investigated how changes in body composition, following consistent exercise, could impact inflammation, which plays a key role in breast cancer recurrence and mortality by promoting tumour progression.

Persistent inflammation not only promotes tumour progression by influencing cell proliferation, survival, invasiveness, and metastasis, but also inhibits immune function. Given that the cancer itself and the side-effects of treatments can elevate levels of inflammatory biomarkers, survivors of breast cancer are at increased risk of cancer progression, recurrence and mortality.

“Strategies are needed to reduce inflammation which may provide a less supportive environment for cancer progression, leading to a lower risk of recurrence and mortality in survivors of breast cancer,” Mr Bettariga said.

The new research found that by reducing fat mass and increasing lean mass, through consistent and persistent exercise, cancer survivors had a better chance at reducing inflammation.

“If we are able to improve body composition, we have a better chance of decreasing inflammation because we are improving lean mass and reducing fat mass, which is responsible for releasing anti and pro-inflammatory markers,” Mr Bettariga said.

Unfortunately, quick fixes to reduce fat mass would not have the same beneficial effects, Mt Bettariga stressed.

“You never want to reduce your weight without exercising, because you need to build or preserve muscle mass and produce these chemicals that you can’t do through just diet alone.”

Story Source:

Materials provided by Canadian Medical Association Journal. Note: Content may be edited for style and length.

Journal Reference:

1. Sophie Li, Marco Solmi, Daniel T. Myran and Nicholas Fabiano. Cannabis and psychosis. CMAJ, 11 August 2025 DOI: 10.1503/cmaj.250659

If the findings of their study are confirmed by other research centers, the biomarker could be the first specific and quantifiable indicator for confirming long COVID. Currently, clinicians confer a diagnosis of long COVID based upon a collection of symptoms that patients develop after SARS-CoV-2 infection.

“If a patient arrives in clinic and they relate the persistence of typical signs and symptoms of long COVID, 12 weeks or more after COVID -19 infection, I give them a presumptive diagnosis, but I don’t have any blood tests or biomarkers to confirm this diagnosis,” said William Stringer, M.D., a Lundquist Institute investigator and senior author on the study.

The study results, reported in the journal Infection, detail the detection of SARS-CoV-2 protein fragments within extracellular vesicles (EVs) — tiny, naturally occurring packages that help cells share proteins, metabolites, and other materials. The researchers collected and analyzed blood samples from 14 patients over 12 weeks of aerobic exercise training (56 samples in all) in a clinical trial led by Stringer in long COVID.

The researchers found 65 distinct protein fragments from SARS-CoV-2 inside the EVs. These fragments come from the virus’s Pp1ab protein, an RNA Replicase enzyme which is key to how the virus copies itself and makes other viral particles. This protein is found uniquely in SARS-CoV-2, and not in uninfected human cells, noted Asghar Abbasi, Ph.D., a Lundquist Institute investigator and first author of the study.

Significantly, the researchers found that these viral peptides were demonstrated in each subject, but not each blood draw, in the EVs of Long COVID patients and were not detected in a separate control group of pre-pandemic EV samples.

These findings add to growing evidence that suggests that SARS-CoV-2 may persist in certain body tissues long after the initial infection. Some groups hypothesize these lingering viral reservoirs could play a role in Long COVID. How the virus reaches tissues without its usual entry points — such as the brain — remains an open question, and may be related to EV particles.

“We thought that maybe if the virus is circulating or moving in the body, we should try to see if EVs are carrying those viral fragments,” Abbasi explained.

This idea became part of an ongoing clinical trial led by Drs. Abbasi and Stringer, which was already studying EVs to see if they are linked to changes in immune function related to exercise and post-exertional malaise, a common symptom in these patients.

“While promising, the molecular signal of the viral peptides within the study samples was observed to be subtle and not consistently detected at every blood collection time point,” said Patrick Pirrotte, Ph.D., associate professor at TGen, director of the Integrated Mass Spectrometry Shared Resource at TGen and City of Hope, and co-senior author of the study. “There’s still a lot to unpack that we don’t know at this point.”

For instance, he added, the researchers don’t know if the exercise itself drives the expression of viral programs intracellularly, and then those viral programs result in proteins that are going to be shed, or if there is a permanent reservoir in those cells, and it’s just a matter of detecting it at a certain time point. Although the identified peptides originated from one of the virus’ largest proteins, the researchers did not detect other comparably large proteins indicative of active viral replication. It’s possible that the peptides contained in the EVs are just molecular “trash” leftover after the formation of new viral proteins.

“We haven’t run [our tests] on people without long COVID symptoms who are currently, or who were, infected with COVID,” said Stringer. “This raises the question: is this just continuing to take out the trash from the COVID infected cell or is this really ongoing replication someplace? I think that’s the mechanistic issue that needs to be resolved in future studies.”

The Pulmonary Education and Research Foundation (PERF) and the UCLA David Geffen School of Medicine (DGSoM)-Ventura County Community Foundation (VCCF) funded this research.