Department of Internal Medicine, Kaohsiung Municipal Ta-Tung Hospital, Kaohsiung Medical University, Kaohsiung, Taiwan.
Division of Nephrology, Kaohsiung Medical University Hospital, Kaohsiung, Taiwan.
PLoS One. 2024 Aug 16;19(8):e0308385. doi: 10.1371/journal.pone.0308385. eCollection 2024.
End-stage kidney disease (ESKD) presents a significant public health challenge, with hemodialysis (HD) remaining one of the most prevalent kidney replacement therapies. Ensuring the longevity and functionality of arteriovenous accesses is challenging for HD patients. Blood flow sound, which contains valuable information, has often been neglected in the past. However, machine learning offers a new approach, leveraging data non-invasively and learning autonomously to match the experience of healthcare professionas. This study aimed to devise a model for detecting arteriovenous grafts (AVGs) stenosis. A smartphone stethoscope was used to record the sound of AVG blood flow at the arterial and venous sides, with each recording lasting one minute. The sound recordings were transformed into mel spectrograms, and a 14-layer convolutional neural network (CNN) was employed to detect stenosis. The CNN comprised six convolution blocks with 3x3 kernel mapping, batch normalization, and rectified linear unit activation function. We applied contrastive learning to train the pre-training audio neural networks model with unlabeled data through self-supervised learning, followed by fine-tuning. In total, 27,406 dialysis session blood flow sounds were documented, including 180 stenosis blood flow sounds. Our proposed framework demonstrated a significant improvement (p<0.05) over training from scratch and a popular pre-trained audio neural networks (PANNs) model, achieving an accuracy of 0.9279, precision of 0.8462, and recall of 0.8077, compared to previous values of 0.8649, 0.7391, and 0.6538. This study illustrates how contrastive learning with unlabeled blood flow sound data can enhance convolutional neural networks for detecting AVG stenosis in HD patients.
终末期肾病 (ESKD) 是一个重大的公共卫生挑战，血液透析 (HD) 仍然是最常见的肾脏替代疗法之一。确保动静脉通路的长期功能对 HD 患者来说是一个挑战。过去，血流声音虽然包含有价值的信息，但往往被忽视。然而，机器学习提供了一种新的方法，利用数据进行非侵入式学习，并自主学习以匹配医疗保健专业人员的经验。本研究旨在设计一种用于检测动静脉移植物 (AVG) 狭窄的模型。使用智能手机听诊器记录 AVG 动脉侧和静脉侧的血流声音，每个记录持续一分钟。声音记录被转换为梅尔频谱图，然后使用 14 层卷积神经网络 (CNN) 检测狭窄。CNN 由六个卷积块组成，具有 3x3 核映射、批量归一化和修正线性单元激活函数。我们应用对比学习，通过自监督学习用未标记的数据对预训练音频神经网络模型进行训练，然后进行微调。总共记录了 27406 次透析治疗血流声音，包括 180 次狭窄血流声音。与之前的 0.8649、0.7391 和 0.6538 相比，我们提出的框架在从零开始训练和流行的预训练音频神经网络 (PANN) 模型上都有显著的提高 (p<0.05)，达到了 0.9279 的准确率、0.8462 的精确率和 0.8077 的召回率。本研究说明了如何使用未标记的血流声音数据进行对比学习，以增强卷积神经网络，从而检测 HD 患者的 AVG 狭窄。
