Studying kinematic linkage of finger joints: estimation of kinematics of distal interphalangeal joints during manipulation

Abstract

The recording of hand kinematics during product manipulation is challenging, and certain degrees of freedom such as distal interphalangeal (DIP) joints are difficult to record owing to limitations of the motion capture systems used. DIP joint kinematics could be estimated by taking advantage of its kinematic linkage with proximal interphalangeal (PIP) and metacarpophalangeal joints. This work analyses this linkage both in free motion conditions and during the performance of 26 activities of daily living. We have studied the appropriateness of different types of linear regressions (several combinations of independent variables and constant coefficients) and sets of data (free motion and manipulation data) to obtain equations to estimate DIP joints kinematics both in free motion and manipulation conditions. Errors that arise when estimating DIP joint angles assuming linear relationships using the equations obtained both from free motion data and from manipulation data are compared for each activity of daily living performed. Estimation using manipulation condition equations implies a lower mean absolute error per task (from 5.87° to 13.67°) than using the free motion ones (from 9° to 17.87°), but it fails to provide accurate estimations when passive extension of DIP joints occurs while PIP is flexed. This work provides evidence showing that estimating DIP joint angles is only recommended when studying free motion or grasps where both joints are highly flexed and when using linear relationships that consider only PIP joint angles.