Abstract #2279

Using deep neural networks to predict RF heating of implanted conductive leads exclusively from implant trajectory and RF coil features

Jasmine Vu^1,2, Bach T Nguyen², Bhumi Bhusal², Justin Baraboo^1,2, Joshua Rosenow³, Ulas Bagci ², Molly G Bright^1,4, and Laleh Golestanirad^1,2

¹Biomedical Engineering, McCormick School of Engineering, Northwestern University, Evanston, IL, United States, ²Radiology, Feinberg School of Medicine, Northwestern University, Chicago, IL, United States, ³Neurosurgery, Feinberg School of Medicine, Northwestern University, Chicago, IL, United States, ⁴Physical Therapy and Human Movement Sciences, Feinberg School of Medicine, Northwestern University, Chicago, IL, United States

Safety risks associated with radiofrequency (RF) heating of tissue around implanted leads limit MRI accessibility for patients with active electronic implants such as those with deep brain stimulation (DBS) devices. RF heating is highly sensitive to the trajectory of the implanted lead, and full-wave electromagnetic simulations are currently the standard method for quantifying RF heating, requiring extensive computational resources and simulation time. Here, we present a promising, fast approach for predicting trajectory-specific maximum local specific absorption rate (SAR) in the tissue around tips of implanted lead models using deep learning.

This abstract and the presentation materials are available to members only; a login is required.

Join Here