Advancing Hepatology with AI: A Systematic Review of Early Detection Models for Hepatitis-Associated Liver Cancer

Background: Despite advancements in health technology, liver cancer remains one of the deadliest forms of cancer and chronic hepatitis B and C viral infections, referred to as HBV and HCV, are major risk factors for the development of HCC. Detecting liver cancer in its early stages is essential to improve cancer patients' results. However, there is a lack of appropriate instruments, such as imaging and biomarkers, which aid traditional cancer detection. The hope raised to counter such diagnosing impediments is the use of artificial intelligence, old-fashioned technologies incorporating development such as machine learning, deep learning and coordinated systems that can narrow the accuracy gap of a cancer diagnosis.  Methods: This systematic review adhered to PRISMA statements and attempts to aggregate published articles on the use of artificial intelligence for diagnosis of primary liver cancer occurring in patients with hepatitis, published in 2020-2024. An extensive search using Booleans was performed in PubMed and Google. Of the 1940 studies found, 50 were appropriate for inclusion. Spine targets, Other AI models, Datasets, Performance metrics and Clinical relevance of the AI implementation were among the main details gathered. Both statistical and narrative approaches accompanied the synthesis of results.  Results: The advancements and applications of AI in diagnosis using AI systems like convolutional neural networks (CNN) and the new hybrid systems are encouraging, with sensitivity and specificity rates consistently over 85% in many cases. Providing images, biomarkers, and other genomic data corroborated this, resulting in high ROC- AUC values. Nonetheless, dataset bias, insufficient real- world applicability, and the requirement for XAI all present significant challenges to using AI in practice.  Conclusion: The use of AI holds great promise in optimizing the early diagnosis of hepatitis-associated HCC by overcoming the challenges posed by conventional methods of diagnosis. To this end, there should be a concentration on increasing the variation in datasets, carrying out extensive research clinical trials, and developing teams spanning different disciplines to allow for easy incorporation into clinical practice. These are promising prospects for enhancing the early detection and treatment of patients suffering from a disease in the Field of hepatology.