Department of Pediatric Neurology, University Children's Hospital, University Medical Centre Ljubljana, Ljubljana, Slovenia.
Department of Catalysis and Chemical Reaction Engineering, National Institute of Chemistry, Ljubljana, Slovenia.
BMJ Paediatr Open. 2024 Aug 30;8(1):e002800. doi: 10.1136/bmjpo-2024-002800.
Cerebral palsy (CP) is a group of neurological disorders with profound implications for children's development. The identification of perinatal risk factors for CP may lead to improved preventive and therapeutic strategies. This study aimed to identify the early predictors of CP using machine learning (ML).
This is a retrospective case-control study, using data from the two population-based databases, the Slovenian National Perinatal Information System and the Slovenian Registry of Cerebral Palsy. Multiple ML algorithms were evaluated to identify the best model for predicting CP.
This is a population-based study of CP and control subjects born into one of Slovenia's 14 maternity wards.
A total of 382 CP cases, born between 2002 and 2017, were identified. Controls were selected at a control-to-case ratio of 3:1, with matched gestational age and birth multiplicity. CP cases with congenital anomalies (=44) were excluded from the analysis. A total of 338 CP cases and 1014 controls were included in the study.
135 variables relating to perinatal and maternal factors.
Receiver operating characteristic (ROC), sensitivity and specificity.
The stochastic gradient boosting ML model (271 cases and 812 controls) demonstrated the highest mean ROC value of 0.81 (mean sensitivity=0.46 and mean specificity=0.95). Using this model with the validation dataset (67 cases and 202 controls) resulted in an area under the ROC curve of 0.77 (mean sensitivity=0.27 and mean specificity=0.94).
Our final ML model using early perinatal factors could not reliably predict CP in our cohort. Future studies should evaluate models with additional factors, such as genetic and neuroimaging data.
脑瘫（CP）是一组具有深远儿童发育影响的神经发育障碍。识别 CP 的围产期危险因素可能会导致改善预防和治疗策略。本研究旨在使用机器学习（ML）识别 CP 的早期预测因子。
这是一项回顾性病例对照研究，使用来自两个基于人群的数据库（斯洛文尼亚国家围产期信息系统和斯洛文尼亚脑瘫登记处）的数据。评估了多种 ML 算法，以确定预测 CP 的最佳模型。
这是一项基于人群的 CP 病例和出生于斯洛文尼亚 14 个产科病房之一的对照研究。
共确定了 382 例 2002 年至 2017 年间出生的 CP 病例。对照组以 3:1 的比例与 CP 病例相匹配，匹配胎龄和出生倍数。从分析中排除了 44 例伴有先天性异常的 CP 病例。共有 338 例 CP 病例和 1014 例对照纳入研究。
135 个与围产期和产妇因素相关的变量。
接收者操作特征（ROC）、敏感性和特异性。
随机梯度增强 ML 模型（271 例病例和 812 例对照）显示最高的平均 ROC 值为 0.81（平均敏感性=0.46，平均特异性=0.95）。使用该模型和验证数据集（67 例病例和 202 例对照），ROC 曲线下面积为 0.77（平均敏感性=0.27，平均特异性=0.94）。
我们使用早期围产期因素的最终 ML 模型不能可靠地预测我们队列中的 CP。未来的研究应评估具有额外因素（如遗传和神经影像学数据）的模型。
