Related Experiment Video For K-means clustering
Updated: Oct 31, 2025
Laparoscopic Choledochal Cyst Excision and Roux-en-Y Choledochojejunostomy in Adults
Published on: February 28, 2025
Cystic Duct Carcinoma: A New Classification System and the Clinicopathological Features of 62 Patients
Lingxi Nan1,2,3, Changcheng Wang1,2,3, Yajie Dai4,5,6
1Department of General Surgery, Zhongshan Hospital, Fudan University, Shanghai, China.
Background:
Cystic duct carcinoma (CDC) is a rare biliary malignancy with a low incidence and poor prognosis. However, the clinical landscape of the disease has not been clarified and no widely applicable classification system has been developed.
Methods:
Sixty-two patients with CDC were included in this retrospective study, and a new classification system was established using imaging data. Blood indices, radiological characteristics, pathological features, surgical procedures, and overall survival data were collected. The efficacy of the new classification in predicting resectability was evaluated using receiver operating characteristic (ROC) curves, and K-means clustering and t-distributed stochastic neighbor embedding were applied to verify the conclusion.
Results:
The pT stage of patients with type II CDC was significantly worse than that of type I. Patients with type II CDC were more likely to experience distant metastasis and invasion of the nervous system, vascular system, and liver. The resectability of patients with type II CDC was significantly worse than that of patients with type I CDC. Patients with type II CDC had worse prognoses. ROC curve analysis and K-means clustering revealed that the new classification could better categorize patients with CDC than currently available systems.
Conclusion:
Patients with type II CDC have significantly worse clinicopathological outcomes. The new classification system has better accuracy in grouping patients with CDC.
