Liver fibrosis is a common outcome of chronic liver disease, characterized by accumulating extracellular matrix proteins such as collagens, glycoproteins, and proteoglycans [1]. This condition, a response to chronic liver injury, represents a critical intermediate stage of liver disease and can result from various underlying conditions, including chronic hepatitis B, non-alcoholic fatty liver disease, autoimmune hepatitis, drug-induced liver injury, and others. While only a minority of patients with liver fibrosis will develop significant cirrhosis, the associated complications, such as ascites, renal failure, hepatic encephalopathy, and variceal bleeding, can significantly impact the management and prognosis of the disease. Currently, no approved antifibrotic therapies exist, but proper treatment in the early stages of liver fibrosis can effectively reverse the condition. Therefore, accurate liver fibrosis detection and staging are essential for disease prognosis and patient management [2,3].

Liver biopsy has historically been the standard for assessing liver fibrosis, but it is an invasive method associated with non-negligible risks of complications. Therefore, reliable clinical non-invasive assessment methods are needed. Although serum biomarkers have widespread clinical applications, their results have been mixed [4]. Transient elastography (TE) measurement of liver stiffness (LSM) is increasingly used for staging liver fibrosis, but it has high failure rates in patients with obesity [5]. Magnetic resonance imaging (MRI) is rapidly gaining popularity and acceptance in clinical practice, especially in quantitative sequences. Among the non-invasive options, diffusion-weighted imaging (DWI) is a very attractive choice. Some DWI models have emerged over the past decades, showing promising results in liver fibrosis assessment [3,6,7], but poor reproducibility and repeatability have limited their clinical application. Nonetheless, research on this technology persisted, and the continuous-time random-walk (CTRW) model, a novel non-Gaussian DWI model based on the CTRW theory, has shown great clinical significance for tumor grading and prognosis evaluation [8,9]. However, few studies have explored its application in assessing and staging liver fibrosis.

This study aimed to evaluate the clinical potential of the CTRW model in staging liver fibrosis and compare its diagnostic performance with conventional DWI and fibrosis biomarkers such as aspartate aminotransferase-to-platelet ratio index (APRI), fibrosis-4 index (FIB-4), and liver stiffness measurement (LSM).