|
| 口腔癌根治术后非计划再次手术的风险因素分析：一项基于机器学习的研究 |
| Analysis of Risk Factors for Unplanned Reoperations After Radical Surgery for Oral Cancer: A Study Based on Machine Learning |
| 投稿时间：2024-06-12 修订日期：2024-07-23 |
| DOI： |
|
 |
| 中文关键词: 口腔癌，非计划再手术，风险因素，Nomogram模型，机器学习 |
| 英文关键词:Oral Cancer, Unplanned Reoperation, Risk Factors, Nomogram Model, Machine Learning |
| 基金项目: |
|
| 摘要点击次数: 65 |
| 全文下载次数: 0 |
| 中文摘要: |
| [目的]：本研究旨在确定口腔癌根治术后非计划再次手术的发生率和风险因素。[方法]：收集2017年1月至2023年1月在本院接受治疗的684例口腔癌患者的资料。根据住院期间是否发生非计划再手术分为再手术组和未再手术组。通过支持向量机（Support Vector Machine, SVM）、随机森林和Lasso回归对再手术的特征因素进行筛选，并采用多因素logistics回归分析独立危险因素。基于危险因素构建Nomogram模型，并评估其预测效能和临床价值。[结果]：再手术组（n=78）患者的性别（P=0.004）、糖尿病史（P=0.009）、现阶段吸烟（P=0.025）、手术部位感染（P<0.001）及皮瓣坏死（P<0.001）发生率均显著高于未手术组（n=606）。计量资料显示，再手术组患者的手术时间（P<0.001）及凝血酶原时间(Prothrombin Time, PT，P=0.041）显著高于未手术组。多因素logistics回归分析表明，手术时间（P<0.001，OR=3.294）、PT（P=0.018，OR=1.850）、血小板计数(Platelet, PLT，P=0.032，OR=2.008）、性别（P=0.014，OR=0.377）、糖尿病史（P=0.008，OR=0.368）、手术部位感染（P=0.005，OR=0.330）及皮瓣坏死（P=0.006，OR=0.267）是影响再手术的独立危险因素。Nomogram模型的曲线下面积（Area under curve, AUC）为0.748，校准曲线与理想曲线几乎重叠，决策曲线（Decision Curve，DCA）曲线显示模型在0~76%之间存在获益率。[结论]：手术时间、PT、PLT、性别、糖尿病史、手术部位感染及皮瓣坏死是口腔癌根治术后非计划再次手术的独立危险因素。通过构建Nomogram模型，可以有效预测患者的再手术风险，有助于临床医生采取针对性措施，减少术后并发症。 |
| 英文摘要: |
| [Objective]: The aim of this study is to determine the incidence and risk factors for unplanned reoperations following radical surgery for oral cancer. [Methods]: We collected data from 684 patients with oral cancer who were treated at our hospital from January 2017 to January 2023. Patients were divided into two groups based on whether they experienced an unplanned reoperation during hospitalization: the reoperation group and the non-reoperation group. Support Vector Machine (SVM), Random Forest, and Lasso regression were used to screen for factors associated with reoperation, and multivariate logistic regression analysis was employed to identify independent risk factors. A Nomogram model was constructed based on these risk factors, and its predictive efficacy and clinical value were assessed. [Results]: Patients in the reoperation group (n=78) had significantly higher rates of gender (P=0.004), history of diabetes (P=0.009), current smoking (P=0.025), surgical site infection (P<0.001), and flap necrosis (P<0.001) compared to the non-reoperation group (n=606). Quantitative data showed that the reoperation group also had significantly longer operation times (P<0.001) and Prothrombin Time (PT) (P=0.041) than the non-reoperation group. Multivariate logistic regression analysis indicated that operation time (P<0.001, OR=3.294), PT (P=0.018, OR=1.850), Platelet (PLT) count (P=0.032, OR=2.008), gender (P=0.014, OR=0.377), history of diabetes (P=0.008, OR=0.368), surgical site infection (P=0.005, OR=0.330), and flap necrosis (P=0.006, OR=0.267) are independent risk factors for reoperation. The Nomogram model had an AUC of 0.748, and the calibration curve almost overlapped with the ideal curve. The Decision Curve Analysis (DCA) showed that the model has a benefit rate between 0~76%. [onclusion]: Operation time, PT, PLT, gender, history of diabetes, surgical site infection, and flap necrosis are independent risk factors for unplanned reoperations after radical surgery for oral cancer. The constructed Nomogram model can effectively predict the risk of reoperation for patients, which may help clinicians to take targeted measures and reduce postoperative complications. |
|
在线阅读
查看/发表评论 下载PDF阅读器 |
|
|
|