A Systematic Analysis of Hand Movement Functionality: Qualitative Classification and Quantitative Investigation of Hand Grasp Behavior

Abstract

Understanding human hand movement functionality is fundamental in neuroscience, robotics, prosthetics, and rehabilitation. People are used to investigate movement functionality separately from qualitative or quantitative perspectives. However, it is still limited to providing an integral framework from both perspectives in a logical manner. In this paper, we provide a systematic framework to qualitatively classify hand movement functionality, build prehensile taxonomy to explore the general influence factors of human prehension, and accordingly design a behavioral experiment to quantitatively understand the hand grasp. In qualitative analysis, two facts are explicitly proposed: (1) the arm and wrist make a vital contribution to hand movement functionality; (2) the relative position (relative position in this paper is defined as the distance between the center of the human wrist and the object center of gravity) is a general influence factor significantly impacting human prehension. In quantitative analysis, the significant influence of three factors, object shape, size, and relative position, is quantitatively demonstrated. Simultaneously considering the impact of relative position, object shape, and size, the prehensile taxonomy and behavioral experiment results presented here should be more representative and complete to understand human grasp functionality. The systematic framework presented here is general and applicable to other body parts, such as wrist, arm, etc. Finally, many potential applications and the limitations are clarified.

Reservoir Computing Model for Human Hand Locomotion Signal Classification

“…Thus, RC algorithms received the best feedback from the machine learning studies and almost no negative feedback. The adjustment of the neuron weights (W in ) of proceeding has a major impact on the ML model and supports a nonlinear dataset [8], [17]. We determined the optimal values of the ML parameters to test and compare the performance in multistep classification.…”

“…a machine tracker via a user interface [5], [6]. The most expressive human gesture is hand locomotion [7], [8]. Initially, the classification of hand gestures in sign language from two-dimensional images was accomplished using a personal computer (PC) [7].…”

“…Due to hand locomotion, these signals have several patterns. AI algorithms such as image processing [20], [21], computer vision [22], principal component analysis (PCA) [8], support vector machine (SVM) [23], artificial neural networks (ANNs) [24], and convolutional neural network (CNNs) [25] can be used for the appropriate analysis and classification of these signals.…”

A straightforward and miniature implementation method of postural synergies to replicate human grasp characteristics accurately and intuitively

“…In order to provide a comprehensive and accurate template reproducing hand grasp postural synergies, we designed an experiment to explore the characteristic movements of human hand joints. Based on our previous study of the qualitative classification and quantitative investigation of hand grasp functionality [29,30], three major factors, including the object shape, size and relative position, were involved in the human grasp experiments. Ten healthy right-handed subjects (24-27 years old, eight men and two women) were asked to grasp six different objects (small and large spheres, cylinders, prisms) in 27 different relative positions (3x × 3y × 3z) between the human hand and the object [29].…”

Design of a Multi-Mode Mechanical Finger Based on Linkage and Tendon Fusion Transmission

“…Some scholars have statistically classified the grasping action types of human hands, such as Feix et al [ 5 ] and Liu et al [ 6 ], who focused on the static and stable grasping of a single hand and, according to (1) opposition type, (2) the virtual finger assignments, (3) type in terms of power, precision, or intermediate grasp, and (4) the position of the thumb, divided the hand’s grasping actions into 33 types. Stival et al [ 7 ], based on the previous classification methods, collected electromyography and kinematic data during human grasping actions to establish the kinematic classification and muscle classification methods for human hand actions.…”