A Hybrid CNN-LSTM Deep Learning Framework for Multi-DOF Control of Upper-Limb Prosthetic Using EMG Signals

Hend ElMohandes; Neamat El Gayar; Adrian Turcanu; Roman Ruff

doi:10.1007/978-981-96-9242-2_26

A Hybrid CNN-LSTM Deep Learning Framework for Multi-DOF Control of Upper-Limb Prosthetic Using EMG Signals

Hend ElMohandes^*
, Neamat El Gayar
, Adrian Turcanu
, Roman Ruff

^*Corresponding author for this work

School of Mathematical & Computer Sciences

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Losing a limb is a life-altering event that significantly affects an individual’s independence and quality of life. Among prosthetic advancements, upper-limb prosthetics have gained greater attention due to their essential role in restoring functionality and autonomy. However, current upper-limb prosthetics are constrained by two primary limitations: restricted ranges of movement and limited simultaneous control. In this paper, we propose a modified CNN-LSTM hybrid architecture designed for continuous computation of four degrees of freedom (DOF), enabling control of the elbow angle (θ), the horizontal (X) and vertical (Y) positions of the wrist joint, and velocity (v) of arm movement. We specifically examine the effect of incorporating historical timesteps on enhancing the decoding performance of these parameters. Our results demonstrate significant enhancements in decoding accuracy when historical timesteps are incorporated surpassing state-of-the-art methods applied on the same dataset and other studies utilizing similar hybrid approaches on different datasets.

Original language	English
Title of host publication	Information System Design. ISDIA 2025
Editors	Vikrant Bhateja, Farhad Oroumchian, Jinshan Tang, Zaid Omar
Publisher	Springer
Pages	355-366
Number of pages	12
ISBN (Electronic)	9789819692422
ISBN (Print)	9789819692415
DOIs	https://doi.org/10.1007/978-981-96-9242-2_26
Publication status	Published - 20 Nov 2025
Event	9th International Conference on Information System Design and Intelligent Applications 2025 - University of Wollongong, Dubai, United Arab Emirates Duration: 3 Jan 2025 → 4 Jan 2025 Conference number: 9 https://isdia.org/

Publication series

Name	Lecture Notes in Networks and Systems
Volume	1537
ISSN (Print)	2367-3370
ISSN (Electronic)	2367-3389

Conference

Conference	9th International Conference on Information System Design and Intelligent Applications 2025
Abbreviated title	ISDIA 2025
Country/Territory	United Arab Emirates
City	Dubai
Period	3/01/25 → 4/01/25
Internet address	https://isdia.org/

Keywords

Continuous decoding
Deep learning
Electromyography
EMG signals
Kinematics
Prosthesis control

ASJC Scopus subject areas

Control and Systems Engineering
Signal Processing
Computer Networks and Communications

Access to Document

10.1007/978-981-96-9242-2_26

Cite this

ElMohandes, H., El Gayar, N., Turcanu, A., & Ruff, R. (2025). A Hybrid CNN-LSTM Deep Learning Framework for Multi-DOF Control of Upper-Limb Prosthetic Using EMG Signals. In V. Bhateja, F. Oroumchian, J. Tang, & Z. Omar (Eds.), Information System Design. ISDIA 2025 (pp. 355-366). (Lecture Notes in Networks and Systems; Vol. 1537). Springer. https://doi.org/10.1007/978-981-96-9242-2_26

@inproceedings{7c012f6794844dc58e47c41d2486438e,

title = "A Hybrid CNN-LSTM Deep Learning Framework for Multi-DOF Control of Upper-Limb Prosthetic Using EMG Signals",

abstract = "Losing a limb is a life-altering event that significantly affects an individual{\textquoteright}s independence and quality of life. Among prosthetic advancements, upper-limb prosthetics have gained greater attention due to their essential role in restoring functionality and autonomy. However, current upper-limb prosthetics are constrained by two primary limitations: restricted ranges of movement and limited simultaneous control. In this paper, we propose a modified CNN-LSTM hybrid architecture designed for continuous computation of four degrees of freedom (DOF), enabling control of the elbow angle (θ), the horizontal (X) and vertical (Y) positions of the wrist joint, and velocity (v) of arm movement. We specifically examine the effect of incorporating historical timesteps on enhancing the decoding performance of these parameters. Our results demonstrate significant enhancements in decoding accuracy when historical timesteps are incorporated surpassing state-of-the-art methods applied on the same dataset and other studies utilizing similar hybrid approaches on different datasets.",

keywords = "Continuous decoding, Deep learning, Electromyography, EMG signals, Kinematics, Prosthesis control",

author = "Hend ElMohandes and \{El Gayar\}, Neamat and Adrian Turcanu and Roman Ruff",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.; 9th International Conference on Information System Design and Intelligent Applications 2025, ISDIA 2025 ; Conference date: 03-01-2025 Through 04-01-2025",

year = "2025",

month = nov,

day = "20",

doi = "10.1007/978-981-96-9242-2\_26",

language = "English",

isbn = "9789819692415",

series = "Lecture Notes in Networks and Systems",

publisher = "Springer",

pages = "355--366",

editor = "Vikrant Bhateja and Farhad Oroumchian and Jinshan Tang and Zaid Omar",

booktitle = "Information System Design. ISDIA 2025",

url = "https://isdia.org/",

}

ElMohandes, H, El Gayar, N, Turcanu, A & Ruff, R 2025, A Hybrid CNN-LSTM Deep Learning Framework for Multi-DOF Control of Upper-Limb Prosthetic Using EMG Signals. in V Bhateja, F Oroumchian, J Tang & Z Omar (eds), Information System Design. ISDIA 2025. Lecture Notes in Networks and Systems, vol. 1537, Springer, pp. 355-366, 9th International Conference on Information System Design and Intelligent Applications 2025, Dubai, United Arab Emirates, 3/01/25. https://doi.org/10.1007/978-981-96-9242-2_26

A Hybrid CNN-LSTM Deep Learning Framework for Multi-DOF Control of Upper-Limb Prosthetic Using EMG Signals. / ElMohandes, Hend; El Gayar, Neamat; Turcanu, Adrian et al.
Information System Design. ISDIA 2025. ed. / Vikrant Bhateja; Farhad Oroumchian; Jinshan Tang; Zaid Omar. Springer, 2025. p. 355-366 (Lecture Notes in Networks and Systems; Vol. 1537).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - A Hybrid CNN-LSTM Deep Learning Framework for Multi-DOF Control of Upper-Limb Prosthetic Using EMG Signals

AU - ElMohandes, Hend

AU - El Gayar, Neamat

AU - Turcanu, Adrian

AU - Ruff, Roman

N1 - Conference code: 9

PY - 2025/11/20

Y1 - 2025/11/20

N2 - Losing a limb is a life-altering event that significantly affects an individual’s independence and quality of life. Among prosthetic advancements, upper-limb prosthetics have gained greater attention due to their essential role in restoring functionality and autonomy. However, current upper-limb prosthetics are constrained by two primary limitations: restricted ranges of movement and limited simultaneous control. In this paper, we propose a modified CNN-LSTM hybrid architecture designed for continuous computation of four degrees of freedom (DOF), enabling control of the elbow angle (θ), the horizontal (X) and vertical (Y) positions of the wrist joint, and velocity (v) of arm movement. We specifically examine the effect of incorporating historical timesteps on enhancing the decoding performance of these parameters. Our results demonstrate significant enhancements in decoding accuracy when historical timesteps are incorporated surpassing state-of-the-art methods applied on the same dataset and other studies utilizing similar hybrid approaches on different datasets.

AB - Losing a limb is a life-altering event that significantly affects an individual’s independence and quality of life. Among prosthetic advancements, upper-limb prosthetics have gained greater attention due to their essential role in restoring functionality and autonomy. However, current upper-limb prosthetics are constrained by two primary limitations: restricted ranges of movement and limited simultaneous control. In this paper, we propose a modified CNN-LSTM hybrid architecture designed for continuous computation of four degrees of freedom (DOF), enabling control of the elbow angle (θ), the horizontal (X) and vertical (Y) positions of the wrist joint, and velocity (v) of arm movement. We specifically examine the effect of incorporating historical timesteps on enhancing the decoding performance of these parameters. Our results demonstrate significant enhancements in decoding accuracy when historical timesteps are incorporated surpassing state-of-the-art methods applied on the same dataset and other studies utilizing similar hybrid approaches on different datasets.

KW - Continuous decoding

KW - Deep learning

KW - Electromyography

KW - EMG signals

KW - Kinematics

KW - Prosthesis control

UR - https://www.scopus.com/pages/publications/105023289869

U2 - 10.1007/978-981-96-9242-2_26

DO - 10.1007/978-981-96-9242-2_26

M3 - Conference contribution

AN - SCOPUS:105023289869

SN - 9789819692415

T3 - Lecture Notes in Networks and Systems

SP - 355

EP - 366

BT - Information System Design. ISDIA 2025

A2 - Bhateja, Vikrant

A2 - Oroumchian, Farhad

A2 - Tang, Jinshan

A2 - Omar, Zaid

PB - Springer

T2 - 9th International Conference on Information System Design and Intelligent Applications 2025

Y2 - 3 January 2025 through 4 January 2025

ER -

A Hybrid CNN-LSTM Deep Learning Framework for Multi-DOF Control of Upper-Limb Prosthetic Using EMG Signals

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this