Real-time control of in vivo thermal ablation using three-dimensional echo decorrelation imaging in swine liver

• Published in: Proceedings of Meetings on Acoustics Vol. 50; no. 1
• Main Authors: Z., Elmira Ghahramani, Grimm, Peter D., Quillin, Ralph C., Patel, Sameer H., Schoenleb, Nicholas S., Weiss, Benjamin E., Saremi, Bahar, Connolly, Benjamin P., Ahmad, Syed A., Shah, Shimul A., Mast, T. Douglas
• Format: Conference Proceeding
• Language: English
• Published: 05.12.2022
• ISSN: 1939-800X
• DOI: 10.1121/2.0001820

In experiments modeling clinical liver tumor treatments, radiofrequency ablation (RFA) of swine liver was controlled using three-dimensional (3D) ultrasound echo decorrelation imaging. Seven ablations were performed in livers of three swine using a clinical RFA system (RITA/Angiodynamics, 50 W, 6-10 min). Paired sequential volumes of beamformed pulse-echo data (inter-frame time <50 ms) were acquired from a Siemens SC2000 scanner with a Z6Ms transesophageal matrix array and transferred via Ethernet to a computer running a custom MATLAB program to compute 3D echo decorrelation images. For controlled trials (N = 2), ablation was automatically ceased when the average cumulative, motion-compensated echo decorrelation within the planned ablation zone exceeded a preset threshold. For trials controlled by the RF generator's default algorithm (uncontrolled trials, N = 5), ablation proceeded for 5 min after the average temperature from three integrated thermocouples reached 105 °C. Following each procedure, the animal's liver was excised, uniformly sectioned, and optically scanned to reconstruct 3D ablation zones. Local ablation prediction was assessed using receiver-operating-characteristic curve analysis, comparing echo decorrelation to local ablation. Ablation volume, ablation rate, and spatially averaged decorrelation were compared for controlled vs. uncontrolled trials. Results indicate promise for control of in vivo thermal ablation using motion-corrected 3D echo decorrelation imaging.