Classification of Individual and Combined Finger Flexions Using Machine Learning Approaches

Hristov, Blagoj; Nadzinski, Gorjan; Latkoska, Vesna Ojleska; Zlatinov, Stefan

doi:10.1109/icca54724.2022.9831952

Cited by 1 publication

(2 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There have been many successful attempts at classifying the movements carried out by users through the use of electromyography, although most fall short due to the use of multiple expensive EMG sensors [4], complex solutions that cannot be easily implemented in an embedded system [5,6] , or a limited array of possible movements [7]. We have already addressed some of these issues in previous publications where we successfully achieved the classification of individual finger flexions using only two-channel electromyography with an F1 score of 91.3% [8], as well as individual and combined finger flexions with an F1 score of 96.6% [9].…”

Section: Related Workmentioning

confidence: 99%

“…The main focus of the paper is the mechanical design of this proposed prosthesis and the different aspects considered while designing the device to be cost effective yet efficient and robust in terms of weight, usability and durability. The signal processing and control aspects have already been covered in [8] and [9], and the discussion on the mechanical considerations presented here represents the next step in the overall design of a low-cost fully functioning forearm prosthesis.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Mechanical Design of a Low-Cost 3d Printed Artificial Forearm Prosthesis With the Ability for Individual Finger Control

Hristov,

Nair,

Rafajlovski

et al. 2023

JEEIT

View full text Add to dashboard Cite

The prosthesis technology has had a renaissance in recent years, although the improved and natural control coupled with fast and fluid movements have led to devices which are very expensive and not attainable by the majority of people who need them. In this paper we present a mechanical design of a low-cost of 3D printed forearm prosthesis which has the ability for individual finger control using a sophisticated software solution by incorporating artificial intelligence. By using two actuators for each finger instead of one we allow for more precise and robust movements of the prosthetic, and by using affordable yet high-torque servomotors we can keep the price low while offering performance similar to much more expensive devices available on the market. The use of 3D printing is also used in order to drive down the cost of the prosthesis, by implementing materials which are inexpensive yet light, strong and durable.

show abstract