Artificial intelligence (AI) is revolutionizing the diagnosis and treatment of biliary diseases, shifting from experience-driven to data-driven approaches. By integrating multi-source data, AI demonstrates significant advantages in early diagnosis, personalized treatment, and prognosis management of biliary diseases. It not only enhances diagnostic accuracy and treatment precision, but also optimizes medical resource allocation and reduces medical costs. The authors systematically elaborate on the current application status and challenges faced by AI in the diagnosis and treatment of biliary diseases. It is believed that the problems, such as data barriers, algorithm limitations, and ethical issues faced in AI clinical applications, need to be solved by establishing a standardized data governance system, developing explainable deep learning models, and improving the medical AI ethical review mechanism. In the future, AI will become a core enabler in the biliary disease diagnosis and treatment ecosystem, driving the transition to knowledgeable healthcare and leading to a new era of precision medicine.

Artificial intelligence (AI) is revolutionizing the diagnosis and treatment of biliary diseases, shifting from experience-driven to data-driven approaches. By integrating multi-source data, AI demonstrates significant advantages in early diagnosis, personalized treatment, and prognosis management of biliary diseases. It not only enhances diagnostic accuracy and treatment precision, but also optimizes medical resource allocation and reduces medical costs. The authors systematically elaborate on the current application status and challenges faced by AI in the diagnosis and treatment of biliary diseases. It is believed that the problems, such as data barriers, algorithm limitations, and ethical issues faced in AI clinical applications, need to be solved by establishing a standardized data governance system, developing explainable deep learning models, and improving the medical AI ethical review mechanism. In the future, AI will become a core enabler in the biliary disease diagnosis and treatment ecosystem, driving the transition to knowledgeable healthcare and leading to a new era of precision medicine.

Objective:To investigate the therapeutic effect and potential molecular mechanisms of cyclin-dependent kinase inhibitor-73 (CDKI-73), the Rab11 inhibitor, on liver fibrosis.Methods:Human LX2 cells were divided into four groups: negative control group, transforming growth factor-β (TGF-β) group, CDKI-73 group and TGF-β+ CDKI-73 group. Fifteen 5-week-old female C57 mice with body weight of (18.04±0.62) g were divided into 3 groups with 5 mice in each group: control group (intraperitoneal injection of olive oil + vehicle gavage), carbon tetrachloride (CCl 4) group (intraperitoneal injection of CCl 4 + vehicle gavage) and CCl 4+ CDKI-73 group (intraperitoneal injection of CCl 4+ CDKI-73 gavage). Another 15 5-week-old female C57 mice with body weight of (18.06±0.34) g were divided into 3 groups with 5 mice in each group: sham operation group (Sham), bile duct ligation (BDL) group + vehicle group (BDL+ vehicle gavage) and bile duct ligation+ CDKI-73 group (BDL+ CDKI-73 gavage). The expression of α-smooth muscle actin (α-SMA) and fibronectin(FN)in LX2 cells were analyzed by Western blot. Masson and Sirius red were used to examine the liver fibrosis after CDKI-73 treatment in vivo. Immunohistochemistry (IHC) was utilized to examine the expression of α-SMA in mice liver. Results:Collagen content assessed by Sirius red and Masson staining and α-SMA expression evaluated by IHC were all increased in CCl 4 group compared with control group ( q=38.47, 24.99, 36.79). Moreover, the collagen content and α-SMA expression in CCl 4 + CDKI-73 treatment group were obviously decreased compared with CCl 4 group ( q=24.72, 14.87, 27.50), and the differences were statistically significant (all P<0.001). Compared with Sham group, collagen content and α-SMA expression in bile duct ligation group were increased ( q=28.23, 41.01, 44.16). Furthermore, in BDL group, after treatment with CDKI-73, the collagen content and α-SMA expression were notably decreased ( q=22.88, 34.31 and 33.97, all P<0.001). Consistent with in vivo results, the relative expression levels of α-SMA and FN protein in TGF-β group were higher than those in TGF-β+ CDKI-73 group (α-SMA: 3.71±0.34 vs. 1.28±0.31; FN: 3.21±0.39 vs. 0.83±0.06, all P<0.001). The mRNA relative expression levels of α-SMA and FN in TGF-β group were higher than those in TGF-β+ CDKI-73 group, and the differences were statistically significant ( P<0.001). However, the relative expression of TGF-β receptor Ⅱ protein in CDKI-73 group was higher than those in negative control group (4.68±0.63 vs. 1.00±0.22, P=0.004). The relative expression level of phosphorylated SMAD2 in TGF-β+ CDKI-73 group was lower than those in TGF-β group (1.67±0.24 vs. 3.99±0.44, P<0.001). Transwell assay showed that 0.5 μmol/L CDKI-73 could effectively inhibit the migration of LX2 cells, and the inhibitory ability became stronger with the increase of CDKI-73 concentration. Conclusion:CDKI-73 can inhibit the activation of hepatic stellate cells and liver fibrosis by inhibiting Rab11-dependent TGF-β signaling pathway both in vivo and in vitro.