Computer Assisted Clinical Medicine, Medical Faculty Mannheim, Heidelberg University, Mannheim 68167, Baden Württemberg, Germany; Mannheim Institute for Intelligent Systems in Medicine, Medical Faculty Mannheim, Heidelberg University, Mannheim 68167, Baden Württemberg, Germany.
Department of Radiology and Nuclear Medicine, University Medical Centre Mannheim, Medical Faculty Mannheim, Heidelberg University, Mannheim 68167, Baden Württemberg, Germany.
Z Med Phys. 2024 May;34(2):330-342. doi: 10.1016/j.zemedi.2023.08.001. Epub 2023 Aug 21.
An accurate prognosis of renal function decline in Autosomal Dominant Polycystic Kidney Disease (ADPKD) is crucial for early intervention. Current biomarkers used are height-adjusted total kidney volume (HtTKV), estimated glomerular filtration rate (eGFR), and patient age. However, manually measuring kidney volume is time-consuming and subject to observer variability. Additionally, incorporating automatically generated features from kidney MRI images, along with conventional biomarkers, can enhance prognostic improvement. To address these issues, we developed two deep-learning algorithms. Firstly, an automated kidney volume segmentation model accurately calculates HtTKV. Secondly, we utilize segmented kidney volumes, predicted HtTKV, age, and baseline eGFR to predict chronic kidney disease (CKD) stages >=3A, >=3B, and a 30% decline in eGFR after 8 years from the baseline visit. Our approach combines a convolutional neural network (CNN) and a multi-layer perceptron (MLP). Our study included 135 subjects and the AUC scores obtained were 0.96, 0.96, and 0.95 for CKD stages >=3A, >=3B, and a 30% decline in eGFR, respectively. Furthermore, our algorithm achieved a Pearson correlation coefficient of 0.81 between predicted and measured eGFR decline. We extended our approach to predict distinct CKD stages after eight years with an AUC of 0.97. The proposed approach has the potential to enhance monitoring and facilitate prognosis in ADPKD patients, even in the early disease stages.
准确预测常染色体显性多囊肾病（ADPKD）的肾功能下降对于早期干预至关重要。目前使用的生物标志物是身高校正的总肾体积（HtTKV）、估算肾小球滤过率（eGFR）和患者年龄。然而，手动测量肾体积既耗时又容易受到观察者变异性的影响。此外，将自动生成的肾脏 MRI 图像特征与常规生物标志物结合使用，可以提高预后的改善效果。为了解决这些问题，我们开发了两种深度学习算法。首先，一种自动的肾脏体积分割模型可以准确地计算 HtTKV。其次，我们利用分割后的肾脏体积、预测的 HtTKV、年龄和基线 eGFR 来预测基线就诊后 8 年内慢性肾脏病（CKD）分期>=3A、>=3B 和 eGFR 下降 30%。我们的方法结合了卷积神经网络（CNN）和多层感知机（MLP）。我们的研究包括 135 名受试者，对于 CKD 分期>=3A、>=3B 和 eGFR 下降 30%，我们获得的 AUC 评分分别为 0.96、0.96 和 0.95。此外，我们的算法在预测和测量的 eGFR 下降之间实现了 0.81 的皮尔逊相关系数。我们将我们的方法扩展到预测 8 年后不同的 CKD 分期，AUC 为 0.97。该方法有望增强 ADPKD 患者的监测和预后，即使在疾病早期阶段。
