Cognitive, Behavioral, and Systems NeuroscienceResearch PaperDecoding the activity of grasping neurons recorded from the ventral premotor area F5 of the macaque monkey
Highlights
▶F5 cortical activity has been fed to recognition algorithms for grip decoding. ▶Distinctive hand postures can be predicted by algorithms with great accuracy. ▶The correct grip can be predicted even before movement onset. ▶F5 neuronal discharge could be used in the control of a prosthetic hand.
Section snippets
Experimental procedures
Properties of the neurons analyzed in this paper have been reported in a previous study (Raos et al., 2006). Single unit activity was recorded from area F5 in the posterior bank of the inferior limb of the arcuate sulcus in three hemispheres (contralateral to the moving forelimb) of two awake monkeys (Macaca nemestrina). The behavioral apparatus and paradigm are summarized in the next section (for more details see (Raos et al., 2006)). All experimental protocols were approved by the
Results
Different algorithms have been tested for the classification of the features extracted from the signals recorded from F5 grasping-related neurons at time intervals spanning from 25% to 100% of movement or object presentation periods, using different sets of objects/grips (original and special, see Fig. 3). In particular, five classifiers have been tested to discriminate either six grips as if each object required a different grip, or clusters of grips (from three to five) according to the
Discussion
The results of the present study show that F5 grasping-related neurons represent a reliable source of information for the implementation of decoding algorithms that could eventually be used for the control of artificial hand grasping. In most cases, few neurons and short window lengths for the extraction of the features were sufficient to achieve a good prediction. The classifier could predict the six grips for the special set of objects and the four grips for the original set of objects.
Acknowledgments
The work described here was partly supported by the EU within the NEUROBOTICS Integrated Project (IST-FET Project 2003-001917 “The fusion of NEUROscience and roBOTICS”).
References (74)
- et al.
Cortical control of grasp in non-human primates
Curr Opin Neurobiol
(2007) - et al.
Development of a lightweight and adaptable multiple-axis hand prosthesis
Med Eng Phys
(2000) - et al.
Patterns of cytochrome oxidase activity in the frontal agranular cortex of the macaque monkey
Behav Brain Res
(1985) - et al.
Cortical projection to hand-arm motor area from post-arcuate area in macaque monkeys: a histological study of retrograde transport of horseradish peroxidase
Neurosci Lett
(1979) - et al.
Frontal lobe inputs to primate motor cortex: evidence for four somatotopically organized “premotor” areas
Brain Res
(1979) - et al.
Hand kinematics during reaching and grasping in the macaque monkey
Behav Brain Res
(2000) - et al.
Brain-controlled interfaces: movement restoration with neural prosthetics
Neuron
(2006) - et al.
The effect of correlated variability on the accuracy of a population code
Neural Comput
(1999) - et al.
Decoding individuated finger movements using volume-constrained neuronal ensembles in the M1 hand area
IEEE Trans Neural Syst Rehabil Eng
(2008) - et al.
Asynchronous decoding of dexterous finger movements using M1 neurons
IEEE Trans Neural Syst Rehabil Eng
(2008)
Correlated firing in macaque visual area MT: time scales and relationship to behavior
J Neurosci
Context-specific grasp movement representation in the macaque anterior intraparietal area
J Neurosci
Decoding M1 neurons during multiple finger movements
J Neurophysiol
Neural networks for pattern recognition
Ventral premotor and inferior parietal cortices make distinct contribution to action organization and intention understanding
Cereb Cortex
Projections of the hand field of the macaque ventral premotor area F5 to the brainstem and spinal cord
J Comp Neurol
Tutorial on support vector machines for pattern recognition
Data Min Knowl Discov
Learning to control a brain-machine interface for reaching and grasping by primates
PLoS Biol
Design of a cybernetic hand for perception and action
Biol Cybern
Facilitation from ventral premotor cortex of primary motor cortex outputs to macaque hand muscles
J Neurophysiol
LIBSVM: a library for support vector machines
A tutorial on v-support vector machines
Appl Stoch Model Bus Ind
Support-vector networks
Mach Learn
PRTools4A Matlab toolbox for pattern recognition
The origin of corticospinal projections from the premotor areas in the frontal lobe
J Neurosci
Frontal lobe inputs to the digit representations of the motor areas on the lateral surface of the hemisphere
J Neurosci
The dorsomedial pathway is not just for reaching: grasping neurons in the medial parieto-occipital cortex of the macaque monkey
J Neurosci
Introduction to statistical pattern recognition
Neurophysiology of prehensionIII. Representation of object features in posterior parietal cortex of the macaque monkey
J Neurophysiol
Neurophysiology of prehensionI. Posterior parietal cortex and object-oriented hand behaviors
J Neurophysiol
Decoding continuous and discrete motor behaviors using motor and premotor cortical ensembles
J Neurophysiol
The science of neural interface systems
Annu Rev Neurosci
Topographic organization of corticospinal projections from the frontal lobe: motor areas on the lateral surface of the hemisphere
J Neurosci
Neuronal ensemble control of prosthetic devices by a human with tetraplegia
Nature
A practical guide to support vector classification
Generalized Bradley-Terry models and multi-class probability estimates
J Mach Learn Res
Intersegmental coordination during reaching at natural visual objects
Cited by (43)
Prediction of Reach Goals in Depth and Direction from the Parietal Cortex
2018, Cell ReportsCitation Excerpt :Naive Bayes methods are a set of supervised learning algorithms based on applying Bayes’ theorem with the “naive” assumption of independence between every pair of features. This technique has been shown to achieve performance closer to optimal compared with other classifiers such as support vector machine (SVM) when analyzing neural data (Carpaneto et al., 2011; Schaffelhofer et al., 2015). In our Python custom scripts, we implemented the module of naive Bayes classifiers proposed by scikit-learn libraries (the statistical formulation can be found at http://scikit-learn.org/stable/modules/naive_bayes.html; Zhang, 2004).
The macaque lateral grasping network: A neural substrate for generating purposeful hand actions
2017, Neuroscience and Biobehavioral ReviewsCitation Excerpt :The selectivity of this deficit of visuomotor transformations for grasping was demonstrated by the observation that reaching and movement time were not clearly affected and that objects could be grasped after tactile exploration. Furthermore, recent modelling studies have shown that AIP and F5 grasping-related activity can be successfully used for decoding appropriate hand–object interactions, suggesting that these two areas represent potentially reliable sources of signals for guiding functional neural interfaces for hand grasping (Carpaneto et al., 2011; Janssen and Scherberger, 2014; Schaffelhofer et al., 2015; Townsend et al., 2011). Area PFG is an architectonically distinct area located in the mid-rostral part of the IPL convexity cortex adjacent to AIP (Fig. 1; Gregoriou et al., 2006).
A machine learning approach to characterize sequential movement-related states in premotor and motor cortices
2022, Journal of NeurophysiologyOffline eeg-based dc motor control for wheelchair application
2021, Lecture Notes in Electrical Engineering