Model based on eight independent variables.

ObjectiveTo identify survival-related risk factors in patients with advanced schistosomiasis, develop a predictive model using integrative approach of LASSO-Cox regression, and construct a nomogram for visualizing the model’s risk prediction framework.MethodsData from 628 advanced schistosomiasis patients treated at Dongzhi Schistosomiasis Hospital between 2019 and 2022 were retrospectively analyzed. LASSO regression was used to select variables associated with survival outcomes, which were subsequently incorporated into a Cox proportional hazards (CPH) model. Internal validation included assessments of discriminative ability (C-index, area under the receiver operating characteristic curve [AUC]), calibration (calibration curves), and clinical utility (decision curve analysis) to evaluate model performance. The final model was visualized via a nomogram depicting the risk prediction algorithm.ResultsLASSO regression identified four independent predictors: carbohydrate antigen 125, hyaluronic acid, ascites grade Ⅱ, and ascites grade Ⅲ. The LASSO-Cox model exhibited strong discriminative performance, with a C-index of 0.886 (SE = 0.025) in the training set and 0.922 (SE = 0.025) in the validation set. Calibration curves showed excellent agreement between predicted and observed survival probabilities, and decision curve analysis confirmed clinical utility across a range of threshold probabilities. A nomogram was developed to translate the model into a user-friendly visual tool for risk stratification.ConclusionsThe constructed nomogram serves as a practical tool for identifying advanced schistosomiasis patients at high mortality risk. Clinicians can leverage this model to tailor individualized follow-up and treatment strategies, potentially improving long-term outcomes by targeting interventions to patients with the greatest need.

研究目的：明确晚期血吸虫病患者的生存相关危险因素，采用LASSO-Cox回归（LASSO-Cox regression）整合分析方法构建预测模型，并搭建列线图（nomogram）以可视化该模型的风险预测框架。研究方法：对2019年至2022年在东至血吸虫病医院接受治疗的628例晚期血吸虫病患者的临床资料进行回顾性分析。采用LASSO回归筛选与生存结局相关的变量，并将其纳入Cox比例风险（CPH）模型。内部验证通过评估区分能力（C指数、受试者工作特征曲线下面积[AUC]）、校准度（校准曲线）及临床效用（决策曲线分析）来评价模型性能。最终通过列线图可视化该风险预测算法模型。研究结果：LASSO回归筛选出4个独立预测因子：糖类抗原125、透明质酸、腹水Ⅱ级及腹水Ⅲ级。LASSO-Cox模型展现出优异的区分性能，训练集的C指数为0.886（标准误=0.025），验证集的C指数为0.922（标准误=0.025）。校准曲线显示预测生存概率与实际观测生存概率拟合度极佳，决策曲线分析证实该模型在一系列阈值概率下均具有临床实用性。本研究搭建的列线图可将该模型转化为便于临床使用的风险分层可视化工具。研究结论：本研究构建的列线图可作为实用工具，用于识别高死亡风险的晚期血吸虫病患者。临床医师可借助该模型制定个体化的随访与治疗方案，通过对高需求患者实施针对性干预，有望改善患者的长期预后。