glove data healthy subjects

Aim of this study was the description of finger movements in right-handed subjects during tactile exploration of a cuboid, a prototypical task of precise handling. The data consists of the time series of 29 sensors integrated in a glove for each hand of 22 subjects. Of the sensors, 16 recorded the bending of metacarpo-phalangeal (MCP) and proximal interphalangeal (PIP) joints of the fingers, MCP and (interphalangeal) IP joint of the thumb, palm arch and carpo-metacarpal (CMC) joint of the thumb, and abduction between fingers. Using principle component analysis we were able to segregate a short action into motor patterns related to successive manipulations of the object. The fraction of variance described by the principal components indicated that salient features of the single motor acts could be described for each hand by three components. Striking in the finger patterns are the prominent and varying roles of MCP and PIP joints of the fingers, and CMC joint of the thumb. An important aspect of the three components is their representation of distinct finger configurations within the same motor act. Analysis of the individual finger time series confirms synchrony during the task and graph analysis establishes high global efficiency of the network of interrelated finger joints. The computation of finger trajectories in one subject exemplifies the workspace of the task, which differs in the right and left hand. The study substantiates finger gaiting as principal mechanism underlying this prototypical task, ubiquitous in daily object shape recognition but described until now only in artificial systems.本研究旨在描述右利手受试者在触觉探索（tactile exploration）长方体过程中的手指运动情况，该任务为一类典型的精细操作原型任务。本数据集包含22名受试者的每只手所佩戴的集成有29个传感器的手套采集的时间序列数据。其中16个传感器用于采集手指的掌指关节（metacarpo-phalangeal, MCP）与近端指间关节（proximal interphalangeal, PIP）、拇指的掌指关节与指间（interphalangeal, IP）关节、掌弓以及拇指腕掌（carpo-metacarpal, CMC）关节的弯曲程度，同时采集手指间的外展幅度数据。借助主成分分析（principle component analysis），我们可将单次短时动作拆解为与物体连续操控相关的运动模式。由主成分所解释的方差占比表明，针对每只手，仅需3个主成分即可描述单次运动动作的显著特征。手指运动模式中较为突出的特点是，手指的MCP与PIP关节以及拇指的CMC关节发挥着显著且多变的作用。这3个主成分的一项重要意义在于，它们可表征同一运动动作内不同的手指姿态。对单根手指时间序列的分析证实了任务执行过程中的手指同步性，而图分析（graph analysis）则证实了相互关联的手指关节网络具备较高的全局效率。对一名受试者的手指运动轨迹进行计算，可直观展现该任务的工作空间，其在左右手间存在差异。本研究证实，手指步态（finger gaiting）是该典型操控任务的核心机制——该任务在日常物体形状识别中普遍存在，但此前仅在人工系统中得到相关描述。