Applications of Artificial Intelligence in liver cancer: a scoping review – part I, screening and diagnosis. (Preprint)

Abstract

Artificial intelligence (AI) has emerged as a powerful tool in various fields, including medicine, offering the potential to revolutionize the way diseases are diagnosed and treated.

This scoping review explores the applications of AI in primary liver cancer, focusing on screening and diagnosis. Liver cancer, particularly hepatocellular carcinoma (HCC) and intrahepatic cholangiocarcinoma (ICC), presents significant challenges due to late-stage diagnoses and limited treatment options.

A systematic review was performed on PubMed, Embase, Scopus, and Web of Science databases including research published between January the 1st 2020 and September the 30th 2023.

AI-driven models have been developed to enhance screening efforts, utilizing machine learning (ML) algorithms trained on clinical, biochemical, and radiological data to identify high-risk patients. These models demonstrate promising results in early HCC detection, especially in populations with chronic hepatitis B virus (HBV) infection or metabolic dysfunction-associated fatty liver disease (MAFLD). Additionally, AI applications in liver imaging, utilizing deep learning (DL) algorithms such as convolutional neural networks (CNN), have shown remarkable accuracy in segmenting and classifying liver lesions on CT and MR images. However, challenges remain in model validation, standardization, and reproducibility, with many studies lacking external validation and consistency in reporting performance metrics. Furthermore, the transition from model development to real-world implementation poses significant hurdles, highlighting the need for a more rigorous and transparent approach in AI model development.

AI-driven models hold immense potential to improve early detection and diagnosis of primary liver cancer, ultimately leading to better patient outcomes. Further research and collaboration are warranted to address the current limitations and facilitate the integration of AI into clinical practice for liver cancer management.