School of Electrical Engineering, Zhengzhou University, Zhengzhou, 450001, China.
Research Center for Intelligent Science and Engineering Technology of TCM, Zhengzhou, 450001, China.
Sci Rep. 2021 Feb 12;11(1):3762. doi: 10.1038/s41598-021-83134-y.
The annotation procedure of pulse wave contour (PWC) is expensive and time-consuming, thereby hindering the formation of large-scale datasets to match the requirements of deep learning. To obtain better results under the condition of few-shot PWC, a small-parameter unit structure and a multi-scale feature-extraction model are proposed. In the small-parameter unit structure, information of adjacent cells is transmitted through state variables. Simultaneously, a forgetting gate is used to update the information and retain long-term dependence of PWC in the form of unit series. The multi-scale feature-extraction model is an integrated model containing three parts. Convolution neural networks are used to extract spatial features of single-period PWC and rhythm features of multi-period PWC. Recursive neural networks are used to retain the long-term dependence features of PWC. Finally, an inference layer is used for classification through extracted features. Classification experiments of cardiovascular diseases are performed on photoplethysmography dataset and continuous non-invasive blood pressure dataset. Results show that the classification accuracy of the multi-scale feature-extraction model on the two datasets respectively can reach 80% and 96%, respectively.
脉搏波轮廓 (PWC) 的注释过程既昂贵又耗时，从而阻碍了大规模数据集的形成，以满足深度学习的要求。为了在少镜头 PWC 的条件下获得更好的结果，提出了一种小参数单元结构和多尺度特征提取模型。在小参数单元结构中，通过状态变量传递相邻单元的信息。同时，使用遗忘门来更新信息，并以单元序列的形式保留 PWC 的长期依赖。多尺度特征提取模型是一个包含三个部分的集成模型。卷积神经网络用于提取单周期 PWC 的空间特征和多周期 PWC 的节律特征。递归神经网络用于保留 PWC 的长期依赖特征。最后，通过提取的特征在推断层进行分类。对光电容积脉搏波数据集和连续无创血压数据集进行心血管疾病的分类实验。结果表明，多尺度特征提取模型在两个数据集上的分类准确率分别可达 80%和 96%。
