Retrospective evaluation of a novel ultrasound-based imaging analysis software for predicting radiofrequency ablation
Masaya Sato1,2, Ryosuke Tateishi2, Yogev Zohar3
1Department of Clinical Laboratory Medicine, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan.
Objective:
This study aimed to introduce and evaluate a novel software-based system, BioTrace, designed for real-time monitoring of thermal ablation tissue damage during image-guided radiofrequency ablation for hepatocellular carcinoma (HCC).
Methods:
BioTrace utilizes a proprietary algorithm to analyze the temporo-spatial behavior of thermal gas bubble activity during ablation, as seen in conventional B-mode ultrasound imaging. Its predictive accuracy was assessed by comparing the ablation zones it predicted with those annotated by radiologists using contrast-enhanced computed tomography (CECT) 24 hours post-treatment, considered the gold standard. The study included 20 liver tumors.
Results:
The median tumor measurement along the major axis was 1.2 cm. The median Dice coefficient, Sensitivity, and Precision between BioTrace and CECT were 0.90, 0.91, and 0.91, respectively. The intraclass correlation showed excellent agreement in volume size between BioTrace and CECT findings (0.98).
Conclusion:
BioTrace effectively generates an ablation damage prediction map based on real-time ultrasound imaging, accurately predicting the ablation zone as confirmed by 24-hour post-procedural CECT. This system has the potential to enhance the safety and efficacy of ablation procedures in clinical settings.
