Learning Multiparametric Biomarkers for Assessing MR-Guided Focused Ultrasound Treatment of Malignant Tumors
https://www.embs.org/tbme/wp-content/uploads/sites/19/2021/04/TBME-00916-2020-Highlight-Image.jpg
170
177
IEEE Transactions on Biomedical Engineering (TBME)
//www.embs.org/tbme/wp-content/uploads/sites/19/2022/06/ieee-tbme-logo2x.png
A significant challenge for noninvasive MR-guided focused ultrasound (MRgFUS) treatment is assessing the viability of treated tissue during and immediately after MRgFUS procedures. Current clinical assessment uses contrast agents that prevent continuing MRgFUS treatment if tumor coverage is inadequate. This work presents a novel, noncontrast, learned multiparametric MR biomarker that can be used during treatment for iterative assessment with inhibiting treatment continuation. Trained using a novel volume-conserving registration algorithm, the presented noncontrast biomarker outperformed the current clinical standard on a VX2 rabbit tumor model. Details on the registration and deep learning model are included.
read more