ECG Segmentation and Overlap Dataset
收藏资源简介:
该数据集由德黑兰大学电气与计算机工程学院的研究团队创建,旨在帮助训练深度学习模型,以准确地将心电图(ECG)图像数字化。数据集包括285个样本,其中100个样本包含人工验证的信号重叠,185个样本没有重叠。数据集分为两个子集:分割数据集和重叠数据集。分割数据集包含单个导联图像及其对应的真实掩模和原始时间序列信号。重叠数据集包含100个信号重叠样本和185个非重叠样本,用于评估模型在处理信号重叠情况下的性能。
This dataset was created by a research team from the School of Electrical and Computer Engineering, University of Tehran, with the objective of training deep learning models to accurately digitize electrocardiogram (ECG) images. The dataset consists of 285 total samples, 100 of which have manually verified signal overlaps, while the other 185 samples have no overlaps. It is divided into two subsets: the segmentation subset and the overlapping subset. The segmentation subset contains single-lead ECG images along with their corresponding ground-truth masks and raw time-series signals. The overlapping subset includes 100 signal-overlap samples and 185 non-overlap samples, designed to evaluate model performance when processing signal overlap scenarios.




