DNE-113 - A Smartphone-based Multi-test Digitized Neurological Examination Dataset

Towards an accessible vision-based exam and documentation solution using a smartphone/tablet device, we conduct a comprehensive multi-test digitized neurological examination (DNE) dataset collection, namely DNE-113. Collected over 113 participants, DNE-113, a multi-test DNE database of finger tapping, finger to finger, forearm roll, stand-up and walk, and facial activation tests. Patients in DNE-113 were diagnosed with Parkinson’s disease (PD) or at least one other neurological (OD) disorder, based on their clinical record. The OD includes subjects experiencing different conditions affecting various regions in the central/peripheral nervous system, or both. In total, a collection of 147 DNE records with 643 videos, from 113 subjects over all five tests forms the DNE-113 dataset. The pose estimation results from OpenPose, MediaPipe, and VideoPose3D are made publicly available.

旨在构建一套易于访问的基于智能手机/平板电脑设备的视觉考试与文档解决方案，我们进行了全面的、多测试数字化神经学检查（DNE）数据集的收集，命名为DNE-113。DNE-113数据集收集了超过113名参与者的数据，包含了一系列的测试，如手指敲击、手指相碰、前臂滚动、站立和行走，以及面部激活测试。DNE-113中的患者被诊断为帕金森病（PD）或至少一种其他神经疾病（OD），这一诊断依据其临床记录。OD包括那些影响中枢/外周神经系统不同区域或两者的不同状况的受试者。总计，DNE-113数据集由147条DNE记录和643个视频组成，涵盖了113名受试者在五种测试形式中的所有测试。OpenPose、MediaPipe和VideoPose3D的姿态估计结果已公开发布。