Abstract

Hepatocellular carcinoma (HCC) is a common cancer worldwide. Recent international guidelines request an identification of the stage and patient background/condition for an appropriate decision for the management direction. Radiomics is a technology based on the quantitative extraction of image characteristics from radiological imaging modalities. Artificial intelligence (AI) algorithms are the principal axis of the radiomics procedure and may provide various results from large data sets beyond conventional techniques. This review article focused on the application of the radiomics-related diagnosis of HCC using radiological imaging (computed tomography, magnetic resonance imaging, and ultrasound (B-mode, contrast-enhanced ultrasound, and elastography)), and discussed the current role, limitation and future of ultrasound. Although the evidence has shown the positive effect of AI-based ultrasound in the prediction of tumor characteristics and malignant potential, posttreatment response and prognosis, there are still a number of issues in the practical management of patients with HCC. It is highly expected that the wide range of applications of AI for ultrasound will support the further improvement of the diagnostic ability of HCC and provide a great benefit to the patients.

Highlights

  • Liver cancer is the sixth most common cancer by incidence and the fourth most common cause of cancer-related mortality worldwide [1]

  • We are in an era of the possible control of viral activities, which is the major factor for hepatocarcinogenesis, Hepatocellular carcinoma (HCC) is still seriously problematic [2]

  • Liver function reserve by Child–Pugh score and degree of tumor progression, evaluated by imaging modalities such as contrast-enhanced computed tomography (CT) and/or magnetic resonance imaging (MRI), are key factors to decide the appropriate treatment from multiple options, such as surgical resection, local ablation (radiofrequency ablation or microwave ablation under ultrasound (US)/CT guidance), percutaneous ethanol injection, transcatheter arterial chemoembolization (TACE), image-guided high-dose-rate brachytherapy, chemotherapy using molecular targeted agents, liver transplantation and best supportive care

Read more

Summary

Introduction

Liver cancer is the sixth most common cancer by incidence and the fourth most common cause of cancer-related mortality worldwide [1]. A specific harmonic mode enables high sensitivity for microbubble detection while being less affected by artefacts compared with the Doppler mode Based on this background, contrast-enhanced US (CEUS) has become popular due to its capacity for stable and real-time observation with the improved detectability of peripheral blood flow under vascular-phase imaging. Changes of morphological and hemodynamic features, depending on the cellular differentiation and variety of malignant grade, are the unique aspect and characteristic features of HCC [2,3,4,5] These are the most important clinical issues of the diagnosis of HCC, which have not been solved by current imaging methods, and the solution by radiomics is highly expected. We reviewed recent studies employing radiomics based on CT and MRI studies for HCC, and discussed the advantages/disadvantages of US in this manner, as well as future directions

Deep Learning
Workflow
Malignant Potential
Microvascular Invasion
Immunotherapy
Radiomics-Based US for the Diagnosis of HCC
Radiomics-Based US for the Diagnosis of Nontumor Liver Disease
10. Summary and Future Perspectives of AI-Based US
Findings
11. Conclusions

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.