Detection and Monitoring of Thermal Lesions Induced by Microwave Ablation Using Ultrasound Imaging and Convolutional Neural Networks.

Microwave ablation (MWA) for cancer treatment is frequently monitored by ultrasound (US) B-mode imaging in the clinic, which often fails due to the low intrinsic contrast between the thermal lesion and normal tissue. Deep learning, especially convolutional neural network (CNN), has shown significant improvements in medical image analysis. Here, we propose and evaluate an US imaging based on a CNN architecture for the detection and monitoring of thermal lesions induced by MWA in porcine livers. Unlike dealing with images in many visual object recognition tasks, US radiofrequency (RF) data backscattered from the ablated region were utilized to capture features related to the thermal lesion. The dataset comprised of 1640 US RF envelope data matrices and their corresponding gross-pathology images, and were utilized for training and testing. After envelope detection, US B-mode, segmentation results based on CNN (SICNN), and modified CNN (SIm-CNN) for US data were simultaneously reconstructed to reveal the suitability for monitoring of MWA. The SICNN and SIm-CNN outperformed B-mode images for the detection and monitoring of MWA-induced thermal lesions. The values of the area under the receiver operating characteristic curve were 0.8728 and 0.8948 for the SICNN and Sim-CNN, respectively, which were both higher than the value of 0.6904 for B-mode images. Ablated regions that were assessed using SIm-CNN showed a good correlation (J 0.8845, r 0.8739, and E 0.410) to gross-pathology images. This study was the first to illustrate that SIm-CNN has the potential to detect and monitor thermal lesions, and may be utilized as an alternative modality for image-guided MWA treatments.

[An in vivo study of ultrasonic monitoring imaging of microwave ablation based on Nakagami statistic parameter].

This paper explored the feasibility of using ultrasonic Nakagami statistic parameter imaging to evaluate the thermal lesion induced by microwave ablation (MWA) in porcine models. In this paper, thermal lesions were induced in livers and kidneys in 5 swines using a clinical MWA system. During this treatment progress, ultrasonic radiofrequency (RF) data were collected. The dynamic changes of Nakagami parameter in the thermal lesion were calculated, and the ultrasonic B-mode images and Nakagami images were reconstructed simultaneously. The contrast-to-noise ratio (CNR) between the thermal lesion and the surrounding normal tissue was calculated over the MWA procedure. After MWA, a bright hyperechoic region appeared in the ultrasonic Nakagami image as an indicator of the thermal lesion and this bright spot enlarged with lesion development during MWA exposure. The mean value of Nakagami parameter in the liver and kidney increased from 0.78 and 0.79 before treatment to 0.91 and 0.92 after treatment, respectively. During MWA exposure, the mean values of CNR calculated from the Nakagami parameter increased from 0.49 to 1.13 in the porcine liver and increased from 0.51 to 0.85 in the kidney, which were both higher than those calculated from the B-mode images. This in vivo study on porcine models suggested that the ultrasonic Nakagami imaging may provide an alternative modality for monitoring MWA treatment.

Ex Vivo and In Vivo Monitoring and Characterization of Thermal Lesions by High-Intensity Focused Ultrasound and Microwave Ablation Using Ultrasonic Nakagami Imaging.

The feasibility of ultrasonic Nakagami imaging to evaluate thermal lesions by high-intensity focused ultrasound and microwave ablation was explored in ex vivo and in vivo liver models. Dynamic changes of the ultrasonic Nakagami parameter in thermal lesions were calculated, and ultrasonic B-mode and Nakagami images were reconstructed simultaneously. The contrast-to-noise ratio (CNR) between thermal lesions and normal tissue was used to estimate the contrast resolution of the monitoring images. After thermal ablation, a bright hyper-echoic region appeared in the ultrasonic B-mode and Nakagami images, identifying the thermal lesion. During thermal ablation, mean values of Nakagami parameter showed an increasing trend from 0.72 to 1.01 for the ex vivo model and 0.54 to 0.72 for the in vivo model. After thermal ablation, mean CNR values of the ultrasonic Nakagami images were 1.29 dB (ex vivo) and 0.80 dB (in vivo), significantly higher ( ) than those for B-mode images. Thermal lesion size, assessed using ultrasonic Nakagami images, shows a good correlation to those obtained from the gross-pathology images (for the ex vivo model: length, = 0.96; width, = 0.90; for the in vivo model: length, = 0.95; width, = 0.85). This preliminary study suggests that ultrasonic Nakagami parameter may have a potential use in evaluating the formation of thermal lesions with better image contrast. Moreover, ultrasonic Nakagami imaging combined with B-mode imaging may be utilized as an alternative modality in developing monitoring systems for image-guided thermal ablation treatments.

In vivo monitoring of microwave ablation in a porcine model using ultrasonic differential attenuation coefficient intercept imaging.

In this study, the feasibility of using ultrasonic differential attenuation coefficient intercept (Δα0) imaging to evaluate thermal lesions induced by microwave ablation (MWA) was explored using an in vivo porcine model. The attenuation coefficient intercept (Δα0 is estimated by subtracting an initial value of Δα0 images. Receiver operating characteristic (ROC) curves and the area under ROC curve (AUC) were employed to statistically assess the predictability of ultrasonic imaging. Ultrasonic Δα0 values were approximately 0.13 dB/cm and 0.16 dB/cm in a normal liver and kidney, respectively, increasing to 2.9 dB/cm and 2.55 dB/cm in ablated regions after MWA. The CNR values of the ultrasonic Δα0 images (0.9 dB and 0.6 dB in the liver and kidney, respectively) were significantly higher (p < 0.05) than the values of B-mode images (0.6 dB and 0.3 dB). The AUC value of the ultrasonic Δα0 image was higher than the B-mode image value, 0.95 compared with 0.88. This in vivo study suggests that ultrasonic Δα0 imaging has the potential to evaluate thermal lesions with high accuracy and better image contrast for monitoring MWA.

Feasibility of Using Ultrasonic Nakagami Imaging for Monitoring Microwave-Induced Thermal Lesion in Ex Vivo Porcine Liver.

The feasibility of using ultrasonic Nakagami imaging to evaluate thermal lesions induced by microwave ablation (MWA) in ex vivo porcine liver was explored. Dynamic changes in echo amplitudes and Nakagami parameters in the region of the MWA-induced thermal lesion, as well as the contrast-to-noise ratio (CNR) between the MWA-induced thermal lesion and the surrounding normal tissue, were calculated simultaneously during the MWA procedure. After MWA exposure, a bright hyper-echoic region appeared in ultrasonic B-mode and Nakagami parameter images as an indicator of the thermal lesion. Mean values of the Nakagami parameter in the thermal lesion region increased to 0.58, 0.71 and 0.91 after 1, 3 and 5 min of MVA. There were no significant differences in envelope amplitudes in the thermal lesion region among ultrasonic B-mode images obtained after different durations of MWA. Unlike ultrasonic B-mode images, Nakagami images were less affected by the shadow effect in monitoring of MWA exposure, and a fairly complete hyper-echoic region was observed in the Nakagami image. The mean value of the Nakagami parameter increased from approximately 0.47 to 0.82 during MWA exposure. At the end of the postablation stage, the mean value of the Nakagami parameter decreased to 0.55 and was higher than that before MWA exposure. CNR values calculated for Nakagami parameter images increased from 0.13 to approximately 0.61 during MWA and then decreased to 0.26 at the end of the post-ablation stage. The corresponding CNR values calculated for ultrasonic B-mode images were 0.24, 0.42 and 0.17. This preliminary study on ex vivo porcine liver suggested that Nakagami imaging have potential use in evaluating the formation of MWA-induced thermal lesions. Further in vivo studies are needed to evaluate the potential application.

Inverse effects of flowing phase-shift nanodroplets and lipid-shelled microbubbles on subsequent cavitation during focused ultrasound exposures.

This paper compared the effects of flowing phase-shift nanodroplets (NDs) and lipid-shelled microbubbles (MBs) on subsequent cavitation during focused ultrasound (FUS) exposures. The cavitation activity was monitored using a passive cavitation detection method as solutions of either phase-shift NDs or lipid-shelled MBs flowed at varying velocities through a 5-mm diameter wall-less vessel in a transparent tissue-mimicking phantom when exposed to FUS. The intensity of cavitation for the phase-shift NDs showed an upward trend with time and cavitation for the lipid-shelled MBs grew to a maximum at the outset of the FUS exposure followed by a trend of decreases when they were static in the vessel. Meanwhile, the increase of cavitation for the phase-shift NDs and decrease of cavitation for the lipid-shelled MBs had slowed down when they flowed through the vessel. During two discrete identical FUS exposures, while the normalized inertial cavitation dose (ICD) value for the lipid-shelled MB solution was higher than that for the saline in the first exposure (p-value <0.05), it decreased to almost the same level in the second exposure. For the phase-shift NDs, the normalized ICD was 0.71 in the first exposure and increased to 0.97 in the second exposure. At a low acoustic power, the normalized ICD values for the lipid-shelled MBs tended to increase with increasing velocities from 5 to 30cm/s (r>0.95). Meanwhile, the normalized ICD value for the phase-shift NDs was 0.182 at a flow velocity of 5cm/s and increased to 0.188 at a flow velocity of 15cm/s. As the flow velocity increased to 20cm/s, the normalized ICD was 0.185 and decreased to 0.178 at a flow velocity of 30cm/s. At high acoustic power, the normalized ICD values for both the lipid-shelled MBs and the phase-shift NDs increased with increasing flow velocities from 5 to 30cm/s (r>0.95). The effects of the flowing phase-shift NDs vaporized into gas bubbles as cavitation nuclei on the subsequent cavitation were inverse to those of the flowing lipid-shelled MBs destroyed after focused ultrasound exposures.