Editorial: Appraising liver fibrosis with eagle eyes

Abstract

Liver fibrosis is the pivotal component of essentially all chronic liver diseases.1 Tremendous efforts have been contributed by many basic and clinical researchers over the last few decades. By understanding the pathological mechanisms of liver fibrosis, exciting developments keep emerging from the basic, diagnostic, prognostic and therapeutic themes. In the current issue of the Journal of Gastroenterology and Hepatology, Zhao et al. lead us through the fascinating development of liver fibrosis research via the bibliometric and visualized analytic insights over the last two decades—nearly 9000 articles with over 45 000 authors from 97 countries had published in more than 1300 journals.2 This important piece of information highlighted the remarkable achievements in this field. This paper brings us an amazing fast-forward view of all these incredible and thrilling advances in understanding liver fibrosis. I found the publications about artificial intelligence (AI) and noninvasive assessments for liver fibrosis, the heated topics in recent years, particularly appealing.3 Noninvasive assessments are also positioned as accurate prognostic for the prediction of various important clinical outcomes, namely, portal hypertension and related complications, hepatocellular carcinoma (HCC), and deaths.4 The importance of the prognostic role of liver fibrosis is further supported by the fact that the most cited original research paper in this field is about liver fibrosis and long-term outcomes in patients with nonalcoholic fatty liver disease.5 AI facilitates the utilization of data from histopathology, clinical parameters, and/or imaging data to predict liver fibrosis with high accuracy. Laboratory-based AI models generally have high applicability as some common laboratory parameters are part of the routine liver panel. On the other hand, AI-assisted histologic scores enable investigators to improve the quality of clinical trials and facilitate comparison across studies. AI models may function as built-in models in the electronic clinical management systems or via some smartphone applications.6 Currently, many AI models are under rigorous evaluations for their ability to assess and monitor disease severity and predict future outcomes. By then, clinicians will have the necessary tools to improve patient care. Moving one step further, the recent fiery debate over ChatGPT-assisted diagnosis in medicine may bring us the future much closer,7 with a great deal of ethical issues to be reconciled.8 With the wide applicability of the important research findings in the field of liver fibrosis, the management of two billion patients with chronic liver diseases worldwide is amidst a revolutionary process.