论文部分内容阅读
目的:开发一种具有较强临床实用性的老年口腔癌患者围术期并发症发生概率评估系统,使得相关并发症发生概率的评估更加直观与高效。方法:根据APACHEⅡ以及POSSUM评分体系,结合临床实际,确定核心数据项目。采用数据挖掘方法,分析各项临床数据之间潜在的逻辑关系,并以CAG为平台,开发评估系统。结果:历时7个月,整理收集并回顾性录入513个病例的临床数据,建立包含49个临床录入项目的 50×513维数据集(数据库)。经过一系列测试后,采用随机森林算法作为核心算法,建立预测评估模型,进而开发出预测系统(软件),为系统设置自我学习功能,并计划添加数据导出与网络化功能。结论:本评估系统在老年口腔癌患者围术期并发症发生概率的预测中具有较高的临床实用性。
OBJECTIVE: To develop a system for assessing the probability of occurrence of perioperative complications in elderly patients with oral cancer, which makes the evaluation of the probability of related complications more intuitive and efficient. Methods: According to APACHE Ⅱ and POSSUM scoring system, combined with clinical practice, to determine the core data items. The data mining method was used to analyze the potential logical relationship between various clinical data and to develop the evaluation system based on CAG. Results: After 7 months, the clinical data of 513 cases were collected and retrospectively collected to establish a 50 × 513 dimension dataset (database) containing 49 clinical entry items. After a series of tests, the random forest algorithm is used as the core algorithm to establish a predictive evaluation model, and then a prediction system (software) is developed to set up a self-learning function for the system and plans to add data export and network functions. Conclusion: This assessment system has high clinical utility in predicting the incidence of perioperative complications in elderly oral cancer patients.