Microwave Ablation Monitoring based on Automatic Segmentation of Ultrasonic Harmonic Envelope Nakagami Parametric Imaging

Abstract

Effective monitoring of the ablation region is important in microwave ablation surgery, and it is the key to the success of the operation. The Nakagami parameter imaging of ultrasonic harmonic envelope signals can realize non-invasive monitoring of the ablation process, but it cannot estimate the ablation center and the ablation region accurately. A Gaussian approximation adaptive threshold segmentation (GATS) method based on ultrasonic harmonic envelope of Nakagami parameter images is proposed to monitor the microwave ablation regions accurately. First, the high-pass filter is used to obtain the harmonic components of the ultrasound echo RF signals. Then the Nakagami shape parameters of harmonic signal envelopes are estimated and Nakagami parameter images are generated by composite window. Nakagami parameter image is processed by Gaussian approximation to estimate the ablation region. Then the anisotropic smoothing is performed on the approximated image, and the adaptive threshold segmentation is used in smoothed image to obtain the estimate ablation region accurately. In order to verify the effectiveness of the GATS method, ultrasonic radio frequency echo signals of microwave ablation were collected from 10 groups of pig liver samples. From the results of the estimated ablation region compares with actual sample pig liver ablation region, the long-axis and short-axis average relative errors are 2.06 %, 3.38 %, 5.67 % and 2.15 %, 4.07 %, 5.36 % respectively. In summary, the estimated ablation region obtained by GATS of P-M algorithm can estimate the ablation region and boundary more accurately.