The object of this study is to develop a model for prediction of lymph node metastasis among intrahepatic cholangiocarcinoma (ICC) patients. Intrahepatic cholangiocarcinoma is the second most common kind of primary liver cancer, accounting for approximately 10%-15%. There is a lack of agreement regarding the necessity of performing lymph node dissection (LND) in patients with ICC. Currently, the percentage of LND is below 50%, and the rate of sufficient LND (≥6) has plummeted to less than 20%. Consequently, a large proportion of patients are unable to acquire LN status, which hinders the following systematic treatment strategies after surgery:. Therefore, our objective is to construct a LN metastasis model utilizing machine learning techniques, including patients' clinical data and pathology information, with the goal of offering a reference for patients who have not undergone LND or have had inadequate LND.

Inclusion Criteria:

• 1.Patients who were confirmed with intrahepatic cholangiocarcinoma 2. curative-intent hepatectomy 3. Without concurrent extrahepatic disease

Exclusion Criteria:

• 1.Patients lacking complete pathology information, 2. Patients who didn't get curative resection 3.Patients with concurrent extrahepatic disease or had missing follow-up data

The data was collected from patients who underwent curative-intent hepatectomy and were diagnosed with ICC pathologically. The data was collected from the hepato-biliary and pancreatic department of West China Hospital, SCU, between the periods of January 2010 to December 2016 and January 2019 to October 2023.

The data that supports the findings of this study is available from the corresponding author upon reasonable request.