Ohashi Mizuki, Ishikawa Yuya, Arai Satoshi, Nagao Tomoharu, Kitaoka Kaori, Nagasu Hajime, Yano Yuichiro, Kashihara Naoki
Shiga University of Medical Science, NCD Epidemiology Research Center, Shiga, Japan.
Department of Information Environment, Yokohama National University Graduate School of Environment and Information Sciences, Yokohama, Japan.
Clin Exp Nephrol. 2025 Jun;29(6):745-752. doi: 10.1007/s10157-024-02616-1. Epub 2025 Jan 15.
Chronic kidney disease (CKD) represents a significant public health challenge, with rates consistently on the rise. Enhancing kidney function prediction could contribute to the early detection, prevention, and management of CKD in clinical practice. We aimed to investigate whether deep learning techniques, especially those suitable for processing missing values, can improve the accuracy of predicting future renal function compared to traditional statistical method, using the Japan Chronic Kidney Disease Database (J-CKD-DB), a nationwide multicenter CKD registry.
From the J-CKD-DB-Ex, a prospective longitudinal study within the J-CKD-DB, we selected individuals who had at least two eGFR measurements recorded between 12 and 20 months apart (n = 22,929 CKD patients). We used the multiple linear regression model as a conventional statistical method, and the Feed Forward Neural Network (FFNN) and Gated Recurrent Unit (GRU)-D (decay) models as deep learning techniques. We compared the prediction accuracies of each model for future eGFR based on the existing data using the root mean square error (RMSE).
The RMSE values were 7.5 for multiple regression analysis, 7.9 for FFNN model, and 7.6 mL/min/1.73 m for GRU-D model. In the subgroup analysis according to CKD stages, lower RMSE values were observed in higher stages for all models.
Our result demonstrate the predictive accuracy of future eGFR based on the existing dataset in the J-CKD-DB-Ex. The accuracy was not improved by applying deep learning techniques compared to conventional statistical methods.
慢性肾脏病（CKD）是一项重大的公共卫生挑战，其发病率持续上升。在临床实践中，提高肾功能预测能力有助于CKD的早期检测、预防和管理。我们旨在研究深度学习技术，尤其是那些适用于处理缺失值的技术，与传统统计方法相比，能否使用日本慢性肾脏病数据库（J-CKD-DB，一个全国性多中心CKD登记系统）提高未来肾功能预测的准确性。
从J-CKD-DB中的前瞻性纵向研究J-CKD-DB-Ex中，我们选择了间隔12至20个月至少有两次估算肾小球滤过率（eGFR）测量值的个体（n = 22929例CKD患者）。我们使用多元线性回归模型作为传统统计方法，以及前馈神经网络（FFNN）和门控循环单元（GRU）-D（衰减）模型作为深度学习技术。我们使用均方根误差（RMSE），基于现有数据比较每个模型对未来eGFR的预测准确性。
多元回归分析的RMSE值为7.5，FFNN模型为7.9，GRU-D模型为7.6 mL/min/1.73m²。在根据CKD分期进行的亚组分析中，所有模型在较高分期观察到较低的RMSE值。
我们的结果证明了基于J-CKD-DB-Ex中现有数据集对未来eGFR的预测准确性。与传统统计方法相比，应用深度学习技术并未提高准确性。
