Regressing grasping using force myography: an exploratory study

Abstract

BackgroundPartial hand amputation forms more than 90% of all upper limb amputations. This amputation has a notable effect on the amputee’s life. To improve the quality of life for partial hand amputees different prosthesis options, including externally-powered prosthesis, have been investigated. The focus of this work is to explore force myography (FMG) as a technique for regressing grasping movement accompanied by wrist position variations. This study can lay the groundwork for a future investigation of FMG as a technique for controlling externally-powered prostheses continuously.MethodsTen able-bodied participants performed three hand movements while their wrist was fixed in one of six predefined positions. The angle between Thumb and Index finger (), and Thumb and Middle finger () were calculated as measures of grasping movements. Two approaches were examined for estimating each angle: (i) one regression model, trained on data from all wrist positions and hand movements; (ii) a classifier that identified the wrist position followed by a separate regression model for each wrist position. The possibility of training the system using a limited number of wrist positions and testing it on all positions was also investigated.ResultsThe first approach had a correlation of determination () of 0.871 for and . Using the second approach and were obtained. The first approach is over two times faster than the second approach while having similar performance; thus the first approach was selected to investigate the effect of the wrist position variations. Training with 6 or 5 wrist positions yielded results which were not statistically significant. A statistically significant decrease in performance resulted when less than five wrist positions were used for training.ConclusionsThe results indicate the potential of FMG to regress grasping movement, accompanied by wrist position variations, with a regression model for each angle. Also, it is necessary to include more than one wrist position in the training phase.Electronic supplementary materialThe online version of this article (10.1186/s12938-018-0593-2) contains supplementary material, which is available to authorized users.

A Review of Force Myography Research and Development

“…These 48 gestures included 16 grasp types, 16 sign language gestures, and 16 finger and hand movements. For continuous actions, Sadeghi et al showed that the angle between index-and-thumb and the one between the middle finger-and-thumb could be accurately predicted using signals extracted near the wrist [33]. In this study, the authors also accounted for different wrist positions while predicting the angles.…”

“…The percentages shown in the figure include some of the work that used multiple classifications algorithms for performance comparison; the percentage is based on the total instance that each method was used and not on the number of publications. As shown in the figure, 37% of the publications use linear discriminant analysis (LDA), 23% use support vector machine (SVM), 15% use artificial neural network (ANN), and the rest use k-nearest neighbor (KNN) [9,31,37], decision tree (DT) [9], deep neural network (DNN) [59,60], extreme learning machine (ELM) [28,55], Gaussian process regression (GPR) [65], hidden Markov model (HMM) [9], random forest (RF) [33,66], and tree bagging (TB) [67].…”

Investigation of Regression Methods for Reduction of Errors Caused by Bending of FSR-Based Pressure Sensing Systems Used for Prosthetic Applications

“…where y k is the expected value of the reading, y k is the predicted value, y k is the mean of expected values, n is the number of observations, and range y is the range of values in observations of expected values. R 2 and RMSE% are commonly used for assessment of performance of regression methods [50]. Based on these outcome measures, one of the regression algorithms was chosen to be used in this study.…”

Investigation of Channel Selection for Gesture Classification for Prosthesis Control Using Force Myography: A Case Study

“…Recent studies have demonstrated FMG to be a promising HMI for upper limb prosthesis control (Cho et al, 2016;Radmand et al, 2016;Ahmadizadeh et al, 2017;Jiang et al, 2017;Sadeghi Chegani and Menon, 2018). FMG monitors changes in volumetric pattern of user's forearm and detects intentions of the user based on these changes (Rasouli et al, 2016).…”