Article
Study of muscle redistribution technique in signal recognition of intelligent bionic hand
Search Medline for
Authors
Published: | February 6, 2020 |
---|
Outline
Text
Objectives/Interrogation: To investigate the value of muscle redistribution technique (MRT) in the recognition of human motion intention by intelligent bionic hand.
Methods: Between December 2016 and April 2017, 3 patients with upper limb amputation were treated with muscle redistribution technique procedure. All patients were males, average age were (42.7±4.5) years. The amputation levels were located at the distal 1/3 of the forearm, midcarpal joint and carpometacarpal joint. In MRT, 4-6 muscles and tendons in the stump of upper limb were transferred and the tendons were anchored in different areas of the skin. When the muscle actively contracted, the tendon pulled the skin resulting in obvious skin deformation in different areas. The evaluating indicators included skin deformation, capacitive signal data and postoperative complications.
Results and Conclusions: All the 3 patients were followed up averaged (5±2) months. There were 14 MRT procedures site. 3 months after surgery, patients were able to actively control the transferred muscle contraction and produce skin deformation. At the final follow-up, the effective deformation rate was 85.7%. Two kinds of classifiers were identified by linear discriminant analysis (LDA) and quadratic discriminant analysis (QDA) respectively. The overall recognition accuracy were 97.27% and 100% respectively, and the recognition accuracy of each action was 100%.
Conclusion MRT procedure can effectively output the motion intention of human body and increase the number and intensity of motion signals. The procedure provided a novel way for better control the intelligent bionic hand.