Just like cardiac pacemakers, implanted neurostimulation devices send electrical pulses to stimulate nerve activity throughout the body. These electrical stimulation devices have been used to treat and control many diseases, including heart disease, epilepsy, depression and rheumatoid arthritis.
But there are many variables that affect the nerve’s response to stimulation, making the development and use of neurostimulation therapies difficult and complex.
Neural engineers at Duke University have designed a computer model that can more easily simulate how nerves respond to electrical stimulation. The model is able to simulate the activity of more than 50,000 nerve fibers, taking the same time as current industry standards to simulate one nerve fiber. Researchers say this new tool will help design more effective and targeted neuromodulation therapies.
The study was published in the journal Nature Communications and the new tool is free to use.
“Many possible adjustments need to be considered to optimize these devices to achieve effective clinical treatment, whether it is changing the amplitude, duration, shape or frequency of pulses, or changing the position of electrodes,” Warren Grill said. Edmund T. Pratt, Distinguished Professor at Duke University’s School of Biomedical Engineering. Pratt, Jr.)。
“Nerve responses are influenced by the anatomical structure and characteristics of the nerves themselves. You have many options for changing your stimulus settings, but it’s hard to know which changes will bring the biggest improvement.”
Engineers have long relied on a platform called “NEURON” to simulate the response of nerve fibers to electrical stimulation. The “MRG” model of nerve fibers was implemented in NEURON and has been widely used in academic research and industry.
Although MRG models are very accurate, the computing power required to simulate neural responses limits their speed, creating bottlenecks that hinder the use of MRG in real-time modeling and slowing down research to improve existing therapies.
To overcome this long-standing obstacle, Dr. Grill, Minhaj Hussain. Nicole “Nikki” Pelot, a student at Grill’s laboratory and research director at the laboratory, developed S-MF (pronounced “smurf”), an alternative to the MRG nerve fiber model. The simulations of the S-MF nerve fiber model cluster run thousands of times faster than the MRG nerve fiber model cluster without sacrificing accuracy or detail.
Unlike the NEURON and MRG models, which run on a CPU (central processing unit), S-MF runs on a GPU (graphics processing unit), a computer chip that can run thousands of calculations in parallel.
“If we model a single fiber, S-MF is not much faster than NEUON,” Hussain said. “But the huge leap is that it takes S-MF to simulate thousands of nerve fibers in the same time as it takes to simulate one MRG nerve fiber. The human vagus nerve alone contains 100,000 nerve fibers, so the efficiency of this new method is very helpful.”
The vagus nerve is a key target for stimulation therapy because it connects the brainstem to most organs of the trunk, including the heart, lungs, pancreas, stomach and liver. Effective stimulation has been shown to safely treat conditions such as drug-resistant epilepsy, depression and heart failure. However, stimulating off-target fibers in the nerve can cause side effects.
The team simplified the representation of nerve fiber anatomy in the model: the MRG model represents different micron anatomical features along the length of the neuron, while the S-MF focuses on key features that initiate and spread neural activity. The team used machine learning methods to define the electrical parameters of the S-MF to ensure accuracy comparable to that of the MRG model.
“Unlike other studies that use alternative methods to speed up simulations, S-MF is accurate across a wide range of neuroanatomy and stimulation parameters,” Nikki Pelot said. “S-MF also retains many details that other simplified methods ignore, which provides important information for designing better therapies.”
The team used S-MF to simultaneously test different stimulation scenarios for thousands of different nerve fibers and quickly determine the best conditions for optimal nerve stimulation. The S-MF’s GPU-based design allows the team to use machine learning optimization techniques that are faster than those available for NEURON-based models.
To demonstrate the power of S-MF and its machine learning optimization, the team predicted stimulation parameters that would only activate neural activity in half of the vagus nerve and render the other half inactive.
The team’s platform quickly and correctly predicted the stimulus levels and patterns that triggered the required response in human and pig vagal nerve models, activating target nerve fibers while avoiding off-target nerve fibers.
Although S-MF was trained to simulate MRG models of myelinated fibers, an important goal of neuromodulation therapy, the team also demonstrated that their platform can be easily adapted to simulate other types of nerve fibers.
They are exploring how to extend their approach to other neuromodulation techniques, including transcranial magnetic stimulation of the brain, which requires modeling of more complex neuroanatomy and multiple types of neurons in the brain.
“Neuroengineering as a field benefits when we can obtain scalable, efficient and anatomically realistic models,” Hussain said. “We hope that as we continue to use this platform, it will tell us more about the design decisions that stimulation therapies should make so that we can achieve the best results.”
For more content, you can click on the link below the video
Thank you for watching this video. If you like it, please subscribe and like it. thank
Original text:https://medicalxpress.com/news/2024-09-optimizing-electrical-therapies-machine.html
More information: Minhaj A. Hussain et al., Efficient modeling and optimization of nerve fiber responses to electrical stimulation, Nature Communications (2024). DOI：10.1038/s41467-024-51709-8
Journal information: Nature Newsletter
Oil tubing: