Parkinson's Disease Tremor Dataset - ALAMEDA
收藏资源简介：
The ALAMEDA_PD_tremor_dataset.csv contains 92 features extracted from raw accelerometer data after pre-processing, 4 tremor-related labels and some other metadata. In total, it includes 99 columns: The first two columns correspond to the start_timestamp and the end_timestamp of the time window from which the respective features have been extracted. The third column corresponds to the subject_id, which is used to uniquely identify PD patients enrolled in the current study. The next 92 columns correspond to features extracted from raw triaxial accelerometer data collected with the GENEActiv smart bracelets throughout 30-min MDS-UPDRS assessment during in-clinic visits, after applying some preprocessing steps. First, the accelerometer signals were band-pass filtered [2.5 Hz, 12.5 Hz] to enable tremor detection. Then, the magnitude and the first principal component of the filtered signals were computed to attenuate the dependency on sensor placement and orientation. Finally, the transformed signals were segmented into time windows of 2048 samples (or 20.48 sec) with 50% overlap. Then, 92 features were extracted in both time and frequency domains. Spectral features were extracted after applying Fast Fourier Transform. The full list of the extracted features is demonstrated in the table below. These features can feed Machine Learning models to predict the presence/absence of PD tremor. The final 4 columns correspond to tremor-related labels (Constancy_of_rest, Kinetic_tremor, Postural_tremor and Rest_tremor). They derive from the respective MDS-UPDRS III annotations, after transforming them to make them suitable for binary classification. More specifically, zero scores remained 0 to indicate the absence of tremor while positive scores were transformed to 1 to indicate the presence of tremor. Each of these columns can be used as a target to be predicted with the help of Machine Learning models.
ALAMEDA_PD_tremor_dataset.csv 数据集包含预处理后从原始加速度计数据中提取的92项特征、4项与震颤相关的标签，以及若干其他元数据。该数据集总计包含99列：前两列为提取对应特征所用时间窗口的起始时间戳（start_timestamp）与结束时间戳（end_timestamp）；第三列为受试者ID，用于唯一标识本研究纳入的帕金森病（Parkinson's Disease, PD）患者；接下来的92列为经过预处理后，从临床就诊期间使用GENEActiv智能手环采集的原始三轴加速度计数据中提取的特征。具体预处理步骤如下：首先，对加速度计信号进行通带范围为[2.5 Hz, 12.5 Hz]的带通滤波以实现震颤检测；随后，计算滤波后信号的模值与第一主成分，以降低对传感器放置位置与朝向的依赖；最后，将转换后的信号分割为含50%重叠的2048个采样点（即20.48秒）的时间窗口。在此基础上，从时域与频域中提取了92项特征，其中频谱特征通过快速傅里叶变换（Fast Fourier Transform, FFT）得到，完整的特征列表见下表。上述特征可输入机器学习（Machine Learning, ML）模型，用于预测帕金森病震颤的有无。最后4列为与震颤相关的标签，分别为静息震颤持续性（Constancy_of_rest）、运动性震颤（Kinetic_tremor）、姿势性震颤（Postural_tremor）与静息性震颤（Rest_tremor）。这些标签源自对应的MDS-UPDRS III评分注释，经转换以适配二分类任务：具体而言，原始评分为0的样本保留为0，代表无震颤；评分为正值的样本被转换为1，代表存在震颤。上述每一列均可作为机器学习模型的预测目标。
