Tsunoda Ryoya, Kume Keitaro, Kagawa Rina, Sanuki Masaru, Kitagawa Hiroyuki, Mase Kaori, Yamagata Kunihiro
Faculty of Medicine, Department of Nephrology, University of Tsukuba, Tsukuba, Japan.
Faculty of Medicine, Department of Clinical Medicine, University of Tsukuba, Tsukuba, Japan.
PLoS One. 2024 Dec 5;19(12):e0312915. doi: 10.1371/journal.pone.0312915. eCollection 2024.
The billing database of the universal healthcare system in Japan potentially includes large-cohort data of patients with immunoglobulin A nephropathy, diagnosis codes aimed at billing should not be directly used for clinical research because of the risk of misdiagnosis. To solve this problem, we aimed to develop a novel method for identifying patients with immunoglobulin A nephropathy from billing data using machine learning. The medical records and bills of 3,743 patients who consulted nephrologists at a single center were extracted. Patients were labeled to have been diagnosed with immunoglobulin A nephropathy through a review of medical records. A manual analysis of the diagnostic accuracy and machine learning was performed. For machine learning, the datasets were preprocessed in three patterns and assigned to the XGBoost program using five-fold cross-validation. Of all the participants, 437 were labeled as having been diagnosed with immunoglobulin A nephropathy. Bill codes for immunoglobulin A nephropathy were provided to approximately half of them. The manually created criteria consisting of the recommended examinations and treatments in the Japanese guidelines for immunoglobulin A nephropathy showed both specificity and sensitivity < 0.8. In contrast, with the receiver operating characteristic curve analysis, the machine learning process yielded area under the curve values over 0.9 with preprocessing from the clinical viewpoint. Applying machine learning technology to a dataset preprocessed from a clinical viewpoint achieved a high performance in detecting patients with immunoglobulin A nephropathy. This methodology contributes to the construction of a disease-specific cohort using big bill data.
日本全民医疗保健系统的计费数据库可能包含大量免疫球蛋白A肾病患者的数据，但由于存在误诊风险，用于计费的诊断代码不应直接用于临床研究。为了解决这个问题，我们旨在开发一种利用机器学习从计费数据中识别免疫球蛋白A肾病患者的新方法。提取了在单一中心咨询肾病专家的3743名患者的病历和账单。通过病历审查将患者标记为已被诊断为免疫球蛋白A肾病。对诊断准确性和机器学习进行了人工分析。对于机器学习，数据集以三种模式进行预处理，并使用五折交叉验证分配给XGBoost程序。在所有参与者中，437人被标记为已被诊断为免疫球蛋白A肾病。其中约一半人获得了免疫球蛋白A肾病的账单代码。由日本免疫球蛋白A肾病指南中推荐的检查和治疗组成的人工制定标准显示特异性和敏感性均<0.8。相比之下，通过受试者工作特征曲线分析，从临床角度进行预处理的机器学习过程产生的曲线下面积值超过0.9。将机器学习技术应用于从临床角度预处理的数据集在检测免疫球蛋白A肾病患者方面具有很高的性能。这种方法有助于利用大量账单数据构建特定疾病队列。
