Dreem Open Dataset-Healthy (DOD-H)

该数据集是对多导睡眠图信号进行睡眠分期的研究数据集。数据集按照不同的受试者被随机划分为训练集、验证集和测试集，比例分别为60%、20%和20%。该数据集包含16个多导睡眠图传感器，采样率为250赫兹，每个信号包含7500个时间步。研究任务是对30秒的多导睡眠图信号进行分类，将其划分为五种睡眠阶段：觉醒、快速眼动（REM）、N1、N2和N3阶段。

This is a research dataset for sleep staging using polysomnography (PSG) signals. The dataset is randomly divided into training, validation, and test sets based on individual subjects, with the corresponding ratios of 60%, 20%, and 20% respectively. This dataset includes 16 polysomnography sensors, with a sampling rate of 250 Hz, and each signal contains 7500 time steps. The research task is to classify 30-second polysomnography signals into five sleep stages: Wake, Rapid Eye Movement (REM), N1, N2, and N3.

搜集汇总

背景与挑战

背景概述

DOD-H是一个包含多通道睡眠信号和专家标注睡眠阶段的开源数据集，主要用于睡眠分期算法的研究和开发。数据集提供了完整的预处理流程和模型训练框架，支持深度学习模型的训练和评估。