Multivariable-Incorporating Super-Resolution Convolutional Neural Network for Transcranial Focused Ultrasound Simulation

Transcranial focused ultrasound (tFUS) has emerged as a new non-invasive brain stimulation (NIBS) modality, with its exquisite ability to reach deep brain areas at a high spatial resolution. Accurate placement of an acoustic focus to a target region of the brain is crucial during tFUS treatment; however, distortion of acoustic wave propagation through the intact skull casts challenges. High-resolution numerical simulation allows for monitoring of the acoustic pressure field in the cranium but also demands extensive computational loads. In this study, we adopt a super-resolution convolutional neural network (SRCNN) technique to enhance the prediction quality of the FUS acoustic pressure field in the targeted brain regions. The training dataset was acquired by numerical simulations performed at low-(1.0 mm) and high-resolutions (0.5 mm) on three ex vivo human calvaria. Five different 3D SRCNN models were trained by using a multivariable dataset, which incorporated information on the acoustic pressure field, wave velocity, and skull computed tomography (CT) images. We achieved an accuracy of 80.87±4.50% in predicting the focal volume with a substantial improvement of 86.91% in computational cost compared to the conventional high-resolution numerical simulation. The results suggest that the method can greatly reduce the simulation time without sacrificing accuracy.