Discrete action control for prosthetic digits

Krasoulis, Agamemnon; Nazarpour, Kianoush

doi:10.1101/2020.03.25.007203

Cited by 4 publications

(8 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this study, we analysed a previously collected dataset comprising data glove measurements 21 , thus we obtained the discrete labels by thresholding the respective velocity profiles. Alternatively, discrete ground truth labels can be acquired by prompting the user to perform imaginary finger movements with specified direction 30 , in a similar fashion that machine learning-based commercial systems are typically calibrated. This feature renders our approach more suitable than regression in a clinical setting, and also makes it suitable for people with bilateral limb deficiency or amputation.…”

Section: Discussionmentioning

confidence: 99%

“…Given that we have introduced a completely novel paradigm for prosthetic digit control, the main purpose of this work was to systematically explore different parameters of the method and lay the groundwork for the subsequent real-time implementation. We report our implementation and evaluation of the proposed algorithm with amputee participants in a separate study 30 .…”

Section: Discussionmentioning

confidence: 99%

“…With regard to the latter, we evaluate the efficacy of a state-of-the-art method for multi-output classification, namely, classifier chains (CC), which takes output dependencies into account when making predictions. The outcomes of our analysis are used to inform the real-time implementation of the algorithm, which we present in a separate study 30 .…”

mentioning

confidence: 99%

See 2 more Smart Citations

Myoelectric digit action decoding with multi-output, multi-class classification: an offline analysis

Krasoulis

Nazarpour

2020

Sci Rep

Self Cite

View full text Add to dashboard Cite

The ultimate goal of machine learning-based myoelectric control is simultaneous and independent control of multiple degrees of freedom (DOFs), including wrist and digit artificial joints. For prosthetic finger control, regression-based methods are typically used to reconstruct position/velocity trajectories from surface electromyogram (EMG) signals. Unfortunately, such methods have thus far met with limited success. In this work, we propose action decoding, a paradigm-shifting approach for independent, multi-digit movement intent prediction based on multi-output, multi-class classification. At each moment in time, our algorithm decodes movement intent for each available DOF into one of three classes: open, close, or stall (i.e., no movement). Despite using a classifier as the decoder, arbitrary hand postures are possible with our approach. We analyse a public dataset previously recorded and published by us, comprising measurements from 10 able-bodied and two transradial amputee participants. We demonstrate the feasibility of using our proposed action decoding paradigm to predict movement action for all five digits as well as rotation of the thumb. We perform a systematic offline analysis by investigating the effect of various algorithmic parameters on decoding performance, such as feature selection and choice of classification algorithm and multi-output strategy. The outcomes of the offline analysis presented in this study will be used to inform the real-time implementation of our algorithm. In the future, we will further evaluate its efficacy with real-time control experiments involving upper-limb amputees.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Myoelectric digit action decoding with multi-output, multi-class classification: an offline analysis

Krasoulis

Nazarpour

2020

Sci Rep

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 99%

“…With regard to the latter, we evaluate the efficacy of a state-of-the-art method for multi-label classification, namely, classifier chains (CCs), which takes output dependencies into account when making predictions. The outcomes of our analysis are used to inform the real-time implementation of the algorithm, which we present in a separate study 29 .…”

mentioning

confidence: 99%

Myoelectric digit action decoding with multi-label, multi-class classification: an offline analysis

Krasoulis

Nazarpour

2020

Preprint

Self Cite

View full text Add to dashboard Cite

The ultimate goal of machine learning-based myoelectric control is simultaneous and independent control of multiple degrees of freedom (DOFs), including wrist and digit artificial joints. For prosthetic finger control, regression-based methods are typically used to reconstruct position/velocity trajectories from surface electromyogram (EMG) signals. Although such methods have produced highly-accurate results in offline analyses, their success in real-time prosthesis control settings has been rather limited. In this work, we propose action decoding, a paradigm-shifting approach for independent, multi-digit movement intent decoding based on multi-label, multi-class classification. At each moment in time, our algorithm classifies movement action for each available DOF into one of three categories: open, close, or stall (i.e., no movement). Despite using a classifier as the decoder, arbitrary hand postures are possible with our approach. We analyse a public dataset previously recorded and published by us, comprising measurements from 10 able-bodied and two transradial amputee participants. We demonstrate the feasibility of using our proposed action decoding paradigm to predict movement action for all five digits as well as rotation of the thumb. We perform a systematic offline analysis by investigating the effect of various algorithmic parameters on decoding performance, such as feature selection and choice of classification algorithm and multi-output strategy. The outcomes of the offline analysis presented in this study will be used to inform the real-time implementation of our algorithm. In the future, we will further evaluate its efficacy with real-time control experiments involving upper-limb amputees.3/11 5/11

show abstract

Co-adaptive myoelectric control for upper limb prostheses

Bañó¹

View full text Add to dashboard Cite

The work of Carles Igual Bañó to carry out this research and elaborate this dissertation has been supported by the Ministerio de Educación, Cultura y Deporte under the FPU Grant FPU15/02870. One visiting research fellowships (EST18/00544) was also funded by the Ministerio de Educación, Cultura y Deporte of Spain.iii Por otro lado, agradecer a mi familia, a mis padres, todo el esfuerzo y dedicación en mi vida universitaria. Su conocimiento de la universidad y del campo académico ha servido para que fuesen un pilar fundamental para cada una de las decisiones a tomar, mi ejemplo a seguir. A su vez, su confianza en la importancia de la tesis me ha ayudado a seguir en momentos críticos donde no han dudado en ofrecerme toda su ayuda y apoyo. También reconocerles todo el cariño durante todos estos años que tanto ha ayudado en los momentos más duros.Y por último, agradecerle a Laura toda su comprensión y ánimo durante estos años tan turbulentos. Estancias, proyectos, estrés, decisiones, momentos muy malos, momentos muy buenos, pero para todos ellos siempre ha estado dispuesta a apoyarme y entenderme las veces que hiciese falta, y no han sido pocas, siendo también siempre la primera en alegrarse por los éxitos. Sin su ayuda este proyecto no habría podido terminarse. vi

show abstract

Discrete action control for prosthetic digits

Cited by 4 publications

References 47 publications

Myoelectric digit action decoding with multi-output, multi-class classification: an offline analysis

Myoelectric digit action decoding with multi-output, multi-class classification: an offline analysis

Myoelectric digit action decoding with multi-label, multi-class classification: an offline analysis

Co-adaptive myoelectric control for upper limb prostheses

Contact Info

Product

Resources

About