Alimohammadi Ehsan, Bagheri Seyed Reza, Moradi Farid, Abdi Alireza, Lawton Michael T
Department of Neurosurgery, Kermanshah University of Medical Sciences, Kermanshah, Iran.
Department of Neurosurgery, Kermanshah University of Medical Sciences, Kermanshah, Iran.
World Neurosurg. 2025 Jan;193:833-841. doi: 10.1016/j.wneu.2024.10.078. Epub 2024 Nov 13.
To assess the efficacy of machine learning models in identifying factors associated with the need for permanent ventricular shunt placement in patients experiencing intracerebral hemorrhage (ICH) who require emergency cerebrospinal fluid (CSF) diversion.
A retrospective review was performed on patients with ICH requiring urgent CSF diversion who were admitted to our facility between July 2009 and May 2023. A binary logistic regression analysis was carried out to determine independent predictors linked to the development of shunt-dependent hydrocephalus following ICH. Five different machine learning models-random forest (RF), support vector machine (SVM), k-nearest neighbor (k-NN), logistic regression (LR), and Adaptive Boosting (AdaBoost)-were utilized to predict the need for permanent shunting in those with spontaneous ICH necessitating emergency CSF diversion. Additionally, RF techniques were applied to identify the factors affecting the need for permanent ventricular shunt placement in these patients.
A total of 578 patients were included in the analysis. Shunt-dependent hydrocephalus occurred in 121 individuals (20.9%). In the multivariate analysis, the Graeb Score, the length of time the external ventricular drain was in place, and an elevated intracranial pressure greater than 30 mm Hg were significant predictors for the need for permanent CSF diversion (P < 0.05). All predictive models showed commendable performance, with RF achieving the highest accuracy (0.921), followed by SVM (0.906), k-NN (0.889), LR (0.881), and AdaBoost (0.823). RF also excelled over the other models in terms of sensitivity and specificity, with a sensitivity of 0.912 and specificity of 0.892. The area under the curve values for RF, SVM, k-NN, LR, and AdaBoost were recorded at 0.903, 0.820, 0.804, 0.801, and 0.798, respectively.
This research demonstrates that machine learning models can effectively predict the need for permanent CSF diversion in patients with ICH who underwent external ventricular drain placement for urgent CSF diversion, offering important prognostic insights that could facilitate early intervention and lead to potential cost reductions.
评估机器学习模型在识别脑出血（ICH）患者中与永久性脑室分流管置入需求相关因素方面的疗效，这些患者需要紧急脑脊液（CSF）引流。
对2009年7月至2023年5月期间入住我院需要紧急CSF引流的ICH患者进行回顾性研究。进行二元逻辑回归分析以确定与ICH后分流依赖性脑积水发生相关的独立预测因素。使用五种不同的机器学习模型——随机森林（RF）、支持向量机（SVM）、k近邻（k-NN）、逻辑回归（LR）和自适应增强（AdaBoost）——来预测自发性ICH且需要紧急CSF引流的患者中永久性分流的需求。此外，应用RF技术识别影响这些患者永久性脑室分流管置入需求的因素。
共有578例患者纳入分析。121例（20.9%）发生分流依赖性脑积水。在多变量分析中，格雷布评分、脑室外引流管放置时间以及颅内压高于30 mmHg是永久性CSF引流需求的显著预测因素（P<0.05）。所有预测模型均表现出良好性能，RF的准确率最高（0.921），其次是SVM（0.906）、k-NN（0.889）、LR（0.881）和AdaBoost（0.823）。RF在敏感性和特异性方面也优于其他模型，敏感性为0.912且特异性为0.892。RF、SVM、k-NN、LR和AdaBoost的曲线下面积值分别记录为0.903、0.820、0.804、0.801和0.798。
本研究表明，机器学习模型可有效预测因紧急CSF引流而接受脑室外引流的ICH患者永久性CSF引流需求，并提供重要的预后见解，有助于早期干预并可能降低成本。
