Artificial intelligence in the detection, characterisation and prediction of hepatocellular carcinoma: a narrative review

Abstract

Hepatocellular carcinoma (HCC) is a significant cause of morbidity and mortality worldwide. Despite significant advancements in detection and treatment of HCC, its management remains a challenge.Artificial intelligence (AI) has played a role in medicine for several decades, however, clinically applicable AIdriven solutions have only started to emerge, due to gradual improvement in sensitivity and specificity of AI, and implementation of convoluted neural networks. A review of the existing literature has been conducted to determine the role of AI in HCC, and three main domains were identified in the search: detection, characterisation and prediction. Implementation of AI models into detection of HCC has immense potential, as AI excels at analysis and integration of large datasets. The use of biomarkers, with the rise of '-omics', can revolutionise the detection of HCC. Tumour characterisation (differentiation between benign masses, HCC, and other malignant tumours, as well as staging and grading) using AI was shown to be superior to classical statistical methods, based on radiological and pathological images. Finally, AI solutions for predicting treatment outcomes and survival emerged in recent years with the potential to shape future HCC guidelines.These AI algorithms based on a combination of clinical data and imaging-extracted features can also support clinical decision making, especially treatment choice. However, AI research on HCC has several limitations, hindering its clinical adoption; small sample size, single-centre data collection, lack of collaboration and transparency, lack of external validation, and model overfitting all results in low generalisability of the results that currently exist. AI has potential to revolutionise detection, characterisation and prediction of HCC, however, for AI solutions to reach widespread clinical adoption, interdisciplinary collaboration is needed, to foster an environment in which AI solutions can be further improved, validated and included in treatment algorithms. In conclusion, AI has a multifaceted role in HCC across all aspects of the disease and its importance can increase in the near future, as more sophisticated technologies emerge.

Hepatocellular carcinoma surveillance after HBsAg seroclearance

“…With the advancement in risk prediction using artificial intelligence and longitudinal clinical data [79], artificial intelligence-based HCC risk models in the future may better estimate the HCC risk among patients with intermediate to high risk and facilitate a risk score-based surveillance strategy. New biomarkers for HCC are also under active investigation, which may further improve the accuracy of the current risk scores [80].…”

Artificial Intelligence in CT and MR Imaging for Oncological Applications

“…Tumor subtyping may further aid the determination of cancer prognosis [39]. Attempts have been made using AI tools to predict survival outcomes in glioblastoma multiforme based on baseline brain MRI [40] as well as to predict response to chemoembolization in hepatocellular carcinoma (based on baseline liver MRI [41]. A comprehensive understanding of the invasive histopathological and molecular approaches, which provide insight into intratumor heterogeneity and the role of advanced MRI imaging in characterizing microstructures, cellularity, physiology, perfusion, and metabolism, is lacking [42,43].…”

The Role of Artificial Intelligence in the Detection and Implementation of Biomarkers for Hepatocellular Carcinoma: Outlook and Opportunities

“…While multiple biomarkers are known for diagnosing liver cancers, advances in technology, especially with the generation of large biological multi-omics datasets and AI algorithms, have expanded the potential for biomarker detection [15,20,21]. Here, we highlight the discovery process of novel biomarkers through AI techniques before diving into the implementation and evaluation of their clinical utility.…”

“…Artificial intelligence (AI) has emerged as a computational technique to identify and evaluate candidate biomarkers for HCC [11][12][13][14]. In general, AI approaches relevant to biomarker detection can be divided into computational search algorithms, machine learning (ML), and deep learning (DL; e.g., convolutional neural networks, CNNs) [15]. Computational search algorithms utilize structured, iterative approaches to evaluate a list of variables, while ML techniques incorporate an inherent feedback mechanism to adjust model parameters during a training period (with subsequent validation during a testing period).…”

“…Random forest analysis, which is a type of supervised ML algorithm, can be used to create an ensemble of decision trees that collectively make predictions, with each decision tree being trained on a random subset of the data. In contrast, deep learning and associated CNNs often implement unsupervised learning, a technique that accepts unlabeled data and classifies cases based on shared features [15,16]. Because unsupervised learning often identifies features that may not be discernable to the human observer, CNNs often require additional interpretation, but may generate unique insights, including variables previously not emphasized in explicit models [16,17].…”