背景:先前的分型方法无法为肝外胆总管囊肿(ECC)的手术复杂性提供预测性见解。本研究旨在通过对成像结果的聚类,建立一种新的ECC分类系统。此外,它旨在比较已确定的ECC类型之间的差异,并评估手术难度的水平。
方法:通过K均值聚类分析对124例患者的影像学数据进行自动分组。根据新分组的特点,进行了纠正和干预,以建立新的分类。人口统计数据,临床表现,手术参数,并发症,再操作,并根据不同类型对预后指标进行分析。还评估了导致手术时间延长的因素。
结果:ECC的新分类系统:类型A(上段),B型(中段),C型(下段),和D型(整个胆管)。合并症(结石或感染)的发生率差异有统计学意义(P=0.000,P=0.002)。此外,术后胆管炎发生率差异有统计学意义(P=0.046).两组手术时间差异有统计学意义(P=0.001)。年龄,BMI>30,分类,合并结石的存在与手术时间延长显著相关(P=0.002,P=0.000,P=0.011,P=0.011)。
结论:结论:我们利用机器学习驱动的聚类分析,创造了一种新颖的肝外胆管扩张类型学.这个分类,结合年龄等因素,联合结石发生,肥胖,显著影响腹腔镜胆总管囊肿手术的复杂性,为改进手术治疗提供有价值的见解。
BACKGROUND: Prior typing methods fail to provide predictive insights into surgical complexities for extrahepatic choledochal cyst (ECC). This study aims to establish a new classification system for ECC through clustering of imaging results. Additionally, it seeks to compare the differences among the identified ECC types and assess the levels of surgical difficulty.
METHODS: The imaging data of 124 patients were automatically grouped through a K-means clustering analysis. According to the characteristics of the new grouping, corrections and interventions were carried out to establish a new classification. Demographic data, clinical presentations, surgical parameters, complications, reoperation, and prognostic indicators were analyzed according to different types. Factors contributing to prolonged surgical time were also evaluated.
RESULTS: A new classification system of ECC: Type A (upper segment), Type B (middle segment), Type C (lower segment), and Type D (entire bile duct). The incidences of comorbidities (calculus or infection) were significantly different (P=0.000, P=0.002). Additionally, variations in the incidence of postoperative cholangitis were statistically significant (P=0.046). The operative time was significantly different between groups (P=0.001). Age, BMI > 30, classification, and the presence of combined stones exhibit a significant association with prolonged operative time (P=0.002, P=0.000, P=0.011, P=0.011).
CONCLUSIONS: In conclusion, our utilization of machine learning-driven cluster analysis has enabled the creation of a novel extrahepatic biliary dilatation typology. This classification, in conjunction with factors like age, combined stone occurrence, and obesity, significantly influences the complexity of laparoscopic choledochal cyst surgery, offering valuable insights for improved surgical treatment.