Takeda Yusuke, Kaneko Yutaka, Sugimoto Masahiko, Yamashita Hidetoshi, Sasaki Ayako, Mitsui Tetsuo
Department of Ophthalmology and Visual Sciences, Yamagata University Faculty of Medicine, Yamagata, Japan.
Yamagata City Institute of Public Health, Yamagata, Japan.
Ophthalmol Sci. 2025 Jan 18;5(4):100715. doi: 10.1016/j.xops.2025.100715. eCollection 2025 Jul-Aug.
To develop nonimaging machine learning models using clinical data from the first screening to predict the occurrence of retinopathy of prematurity (ROP).
This multicenter regional study was conducted in Yamagata Prefecture, Japan.
We collected clinical data of neonates born between October 2016 and September 2018 and screened in 4 neonatal care units.
The 35 variables available at the first screening were used as possible predictors to develop a decision tree, a random forest, a gradient-boosted tree, a neural network, and a Naive Bayes model. Parameter tuning was performed using a 10-fold cross-validation. This process was repeated 200 times using different random seeds for data partitioning.
The target outcome was the final ROP outcome (i.e., the development of any stage of ROP during hospitalization).
Of the 215 neonates screened, 43 (20.0%) developed ROP. The median gestational age was 31.4 (interquartile range: 28.1-33.4) weeks, and the median birth weight was 1502 (interquartile range: 967-1823) g. The mean 200-iteration area under the receiver operating characteristic curve (AUC-ROC), accuracy, sensitivity, and specificity of the random forest model were 0.93 (95% confidence interval [CI] 0.83-0.99), 90.1% (95% CI 84.1-95.2), 95.7% (95% CI 88.2-100), and 66.0% (95% CI 41.7-91.7), respectively. The mean 200-iteration AUC-ROC, accuracy, sensitivity, and specificity of the Naive Bayes model were 0.94 (95% CI 0.86-0.99), 90.6% (95% CI 84.1-96.8), 94.6% (95% CI 86.3-100), and 73.6% (95% CI 50.0-91.7), respectively.
Nonimaging machine learning methods have shown high performance in predicting the occurrence of ROP. These models can be beneficial when a fundus camera cannot capture images due to eye opacity and for hospitals that lack pediatric fundus cameras.
The author(s) have no proprietary or commercial interest in any materials discussed in this article.
利用首次筛查的临床数据开发非成像机器学习模型，以预测早产儿视网膜病变（ROP）的发生。
本多中心区域研究在日本山形县进行。
我们收集了2016年10月至2018年9月出生并在4个新生儿护理单位接受筛查的新生儿的临床数据。
将首次筛查时可用的35个变量用作可能的预测因子，以开发决策树、随机森林、梯度提升树、神经网络和朴素贝叶斯模型。使用10折交叉验证进行参数调整。使用不同的随机种子进行数据划分，此过程重复200次。
目标结果是最终的ROP结果（即住院期间任何阶段ROP的发生情况）。
在筛查的215名新生儿中，43名（20.0%）发生了ROP。中位胎龄为31.4（四分位间距：28.1 - 33.4）周，中位出生体重为1502（四分位间距：967 - 1823）g。随机森林模型在200次迭代下的平均受试者工作特征曲线下面积（AUC - ROC）、准确率、敏感性和特异性分别为0.93（95%置信区间[CI] 0.83 - 0.99）、90.1%（95% CI 84.1 - 95.2）、95.7%（95% CI 88.2 - 100）和66.0%（95% CI 41.7 - 91.7）。朴素贝叶斯模型在200次迭代下的平均AUC - ROC、准确率、敏感性和特异性分别为0.94（95% CI 0.86 - 0.99）、90.6%（95% CI 84.1 - 96.8）、94.6%（95% CI 86.3 - 100）和73.6%（95% CI 50.0 - 91.7）。
非成像机器学习方法在预测ROP的发生方面表现出高性能。当眼底相机因眼睛混浊无法拍摄图像时，以及对于缺乏儿科眼底相机的医院，这些模型可能会有所帮助。
作者对本文讨论的任何材料均无专有或商业利益。
