Hemodialysis Room, Department of Renal Endocrinology, Anhui Lujiang People's Hospital, Lujiang 231501, China.
Department of Renal Endocrinology, Anhui Lujiang People's Hospital, Lujiang 231501, China.
Comput Math Methods Med. 2022 Jul 25;2022:8752826. doi: 10.1155/2022/8752826. eCollection 2022.
A predictive model was established based on logistic regression and XGBoost algorithm to investigate the factors related to postoperative hypocalcemia in patients with secondary hyperparathyroidism (SHPT).
A total of 60 SHPT patients who underwent parathyroidectomy (PTX) in our hospital were retrospectively enrolled. All patients were randomly divided into a training set ( = 42) and a test set ( = 18). The clinical data of the patients were analyzed, including gender, age, dialysis time, body mass, and several preoperative biochemical indicators. The multivariate logistic regression and XGBoost algorithm models were used to analyze the independent risk factors for severe postoperative hypocalcemia (SH). The forecasting efficiency of the two prediction models is analyzed.
Multivariate logistic regression analysis showed that body mass (OR = 1.203, = 0.032), age (OR = 1.214, = 0.035), preoperative PTH (OR = 1.026, = 0.043), preoperative Ca (OR = 1.062, = 0.025), and preoperative ALP (OR = 1.031, = 0.027) were positively correlated with postoperative SH. The top three important features of XGBoost algorithm prediction model were preoperative Ca, preoperative PTH, and preoperative ALP. The area under the curve of the logistic regression and XGBoost algorithm model in the test set was 0.734 (95% CI: 0.5950.872) and 0.827 (95% : 0.7220.932), respectively.
The predictive models based on the logistic regression and XGBoost algorithm model can predict the occurrence of postoperative SH.
基于逻辑回归和 XGBoost 算法建立预测模型，探讨继发性甲状旁腺功能亢进症（SHPT）患者术后低钙血症相关因素。
回顾性纳入我院行甲状旁腺切除术（PTX）的 60 例 SHPT 患者，将所有患者随机分为训练集（n=42）和测试集（n=18）。分析患者的临床资料，包括性别、年龄、透析时间、体重及术前多项生化指标。采用多因素逻辑回归和 XGBoost 算法模型分析严重术后低钙血症（SH）的独立危险因素，分析两种预测模型的预测效能。
多因素逻辑回归分析显示，体重（OR=1.203， =0.032）、年龄（OR=1.214， =0.035）、术前 PTH（OR=1.026， =0.043）、术前 Ca（OR=1.062， =0.025）和术前 ALP（OR=1.031， =0.027）与术后 SH 呈正相关。XGBoost 算法预测模型中前 3 位重要特征为术前 Ca、术前 PTH 和术前 ALP。逻辑回归和 XGBoost 算法模型在测试集的曲线下面积分别为 0.734（95%CI：0.5950.872）和 0.827（95%CI：0.7220.932）。
基于逻辑回归和 XGBoost 算法模型的预测模型可预测术后 SH 的发生。
