Efficacy of fusion imaging for immediate post‐ablation assessment of malignant liver neoplasms: A systematic review

Abstract

Background: Percutaneous thermal ablation has become the preferred therapeutic treatment option for liver cancers that cannot be resected. Since ablative zone tissue changes over time, it becomes challenging to determine therapy effectiveness over an extended period. Thus, an immediate post-procedural evaluation of the ablation zone is crucial, as it could influence the need for a second-look treatment or follow-up plan. Assessing treatment response immediately after ablation is essential to attain favorable outcomes. This study examines the efficacy of image fusion strategies immediately post-ablation in liver neoplasms to determine therapeutic response.Methodology: A comprehensive systematic search using PRISMA methodology was conducted using EMBASE, MEDLINE (via PUBMED), and Cochrane Library Central Registry electronic databases to identify articles that assessed the immediate post-ablation response in malignant hepatic tumors with fusion imaging (FI) systems. The data were retrieved on relevant clinical characteristics, including population demographics, pre-intervention clinical history, lesion characteristics, and intervention type. For the outcome metrics, variables such as average fusion time, intervention metrics, technical success rate, ablative safety margin, supplementary ablation rate, technical efficacy rate, LTP rates, and reported complications were extracted.Results: Twenty-two studies were included for review after fulfilling the study eligibility criteria. FI's immediate technical success rate ranged from 81.3% to 100% in 17/22 studies. In 16/22 studies, the ablative safety margin was assessed immediately after ablation. Supplementary ablation was performed in 9 studies following immediate evaluation by FI. In 15/22 studies, the technical effectiveness rates during the first follow-up varied from 89.3% to 100%.

Unveiling the future of breast cancer assessment: a critical review on generative adversarial networks in elastography ultrasound

“…Deep learning models have achieved notable success in classifying, segmenting, and detecting relevant ROI in medical images and other modalities of data (2,(29)(30)(31)(32)(33). Recently, neural networks have been employed to upscale low-resolution medical images, transform medical imaging modalities, enhance visualization, and improve diagnostic accuracy.…”

Ensemble learning enhances the precision of preliminary detection of primary hepatocellular carcinoma based on serological and demographic indices

“…Within this plethora of research endeavors, precise segmentation models tailored specifically for liver US and CT images abound (12)(13)(14). These highly effective segmentation techniques have played an instrumental role in providing substantial support for clinical surgical treatments, aiding in intervention decision-making, and enhancing postoperative evaluations for HCC (15)(16)(17). Moreover, the application of AI in the medical domain extends beyond imaging.…”