Deep Learning Approach Enables Radiation-Free TFUS Treatment Planning with Synthetic CT

Abstract

PURPOSE: Acoustic simulation has been widely used in the field of transcranial focused ultrasound (tFUS) to predict the acoustic field in the cranial cavity. Previous efforts to ensure accurate simulation have relied on CT scans to derive the acoustic properties of the skull and integrate them into the simulations. However, reliance on CT imaging poses inherent risks of radiation exposure to patients, potentially increasing the risk of cancer. This study aims to explore the feasibility of using deep learning-based synthetic CT (sCT) generated from commonly used T1-weighted MRI (T1w MRI) for tFUS acoustic simulation.