Supercomputer model helps explain electrical shock

CI Beagle shock header

ZZT, ow! That old extension cord you just reached for behind the entertainment center just delivered a painful surprise: an electric shock to the fingers of your right hand. Despite a quick reflexive retraction, you feel a painful tingling in your fingers and notice small burn marks where your fingers touched the frayed cord. As a precaution you take a trip to the ER, where the doctor prescribes some antibiotic cream for the burn and says the tingling should go away on its own.

For a mild household shock, this treatment is typically enough. But for more severe shocks, and even some lower voltage contacts, physicians and scientists remain in the dark about the extent of the damage caused by electricity. More than just a simple burn, electric shocks can do harm through both high temperature and the passage of an electric field through the precisely charged environment of a human body. With a new model of electric shock injury run on the CI’s Beagle supercomputer, University of Chicago Medicine’s Raphael Lee hopes to understand both of these harmful effects, inspiring new treatments and protective gear.

The human body is an electric machine, using propagating potentials to power the nervous system and depending on many proteins sensitive to voltage for proper folding and function. An electric shock, ranging from a finger in an outlet to a lightning strike, produces a powerful electric field that travels through the body in milliseconds, disrupting all of these finely-tuned processes and structures, perhaps permanently. To computationally simulate this damage, as well as the injury caused by the heat of the shock, requires a detailed anatomical model of the different tissues of the body, and the power to calculate physics of both temperature and electrical effects.

“This is truly a multi-physical, multi-scale, non-linear problem,” said Lee, the Paul and Allene Russell Professor of Surgery, Medicine, Organismal Biology and Anatomy and Fellow at the Institute for Molecular Engineering. “If you want to computationally simulate the extent to which tissues are damaged during electrical shock, that would require an extensive computational power.”

Since the mid-80s, Lee’s group has been working on such a model, first coding a simulation program from scratch on an early, 32-bit VAXstation II. On that machine, a single simulation required 4-5 months to solve the linearized field problems alone for a layered 2-D mesh model of the human arm. But it wasn’t until now, with the 200-teraflop Beagle supercomputer, that the team could simultaneously simulate both the heat and electrical damage accumulation during and after an electrical shock, using a full, 1mm-resolution anatomical model containing all major tissues with their respective material properties.

That level of detail is needed because the tissue damage from supraphysiological electric forces and temperatures are not independent — damage accumulation alters tissue material properties such that the induced scalar potentials are altered during the electrical shock.

“It’s a daunting problem,” Lee said. “You need a machine like the Beagle to have a shot at it.”

The research team, which included post-doctoral researcher Paul Liu, PhD and Beagle user support specialist Ana Marija Sokovic, PhD used a 3-D computational mesh model of human anatomy and the StarCCM software for computational fluid dynamics. As a pilot, they simulated the damage from a one second, 7200-volt electrical shock on an arm — allowing the team to compare and calibrate results from this new model to previous arm-shock simulations performed in the past.

Now that those early experiments are finished, the group hopes to scale up to full-body simulations and tests of the short and long-term consequences of different voltages. They also plan to couple continuum models of scalar potentials and chemical reaction thermodynamics to predict injury dynamics. Some patients who have mild shocks and early symptoms can develop more severe issues over time, such as loss of balance or coordination and pain. The modeling experiments will help physicians better diagnose the severity of electric shock, Lee said, allowing for more specific treatment and early prognoses for patients.

“With burns, you have first, second, or third degree, but right now that kind of stratification doesn’t really exist for electric shock,” Lee said. “Based on learning from these models, we can generate more precise injury intervention protocols that might be effective immediately post-injury.”

The model will also be used to study new forms of protective gear against electric shocks, which could be worn by utility workers exposed to high voltages. In addition, Lee expects, the new model will be of particular importance for a more exotic occupation: predicting harmful stresses experienced by astronauts during space travel. Lee’s group previously worked with NASA to test the effects of electrical shock during extravehicular activity around the International Space Station.

“We want to anticipate and understand what the factors are, so that one can prepare for when we consider traveling to other planets,” Lee said.

This story was originally published by the Computation Institute.

About Rob Mitchum (518 Articles)
Rob Mitchum is communications manager at the Computation Institute, a joint initiative between The University of Chicago and Argonne National Laboratory.
%d bloggers like this: