Using HMM to Model Neural Dynamics and Decode Useful Signals for Neuroprosthetic Control

Diomedi, Stefano; F, Vaccari; Hadjidimitrakis, Kostas; Fattori, Patrizia

doi:10.1007/978-3-030-99142-5_3

Cited by 2 publications

(3 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Following the consistency analysis approach used for V6A and PEc ( Diomedi et al, 2021 ), we identified the optimal number of hidden neural states that may be detected in area PE considering data that spanned from −1,000 ms to +1,000 ms around movement onset. Note that we chose the same feedforward linear topology for the model already used in our previous studies ( Diomedi et al, 2021 , 2022 ), since it seemed the most appropriate given the sequential nature of the reaching task ( Kemere et al, 2008 ). In brief, at each new bin, the Markov process could only remain in the same state as that of the previous bin or shift to the next state of the chain.…”

Section: Resultsmentioning

confidence: 99%

Similar neural states, but dissimilar decoding patterns for motor control in parietal cortex

Vaccari,

Diomedi,

De Vitis

et al. 2024

Network Neuroscience

Self Cite

View full text Add to dashboard Cite

Discrete neural states are associated with reaching movements across the frontoparietal network. Here, the Hidden Markov Model (HMM) applied to spiking activity of the somato-motor parietal area PE revealed a sequence of states similar to those of the contiguous visuo-motor areas PEc and V6A. Using a coupled clustering and decoding approach, we proved that these neural states carried spatio-temporal information regarding behaviour in all three posterior parietal areas. However, comparing decoding accuracy, PE was less informative than V6A and PEc. In addition, V6A outperformed PEc in target inference, indicating functional differences among the parietal areas. To check the consistency of these differences, we used both a supervised and an unsupervised variant of the HMM, and compared its performance with two more common classifiers, Support Vector Machine and Long-Short Term Memory. The differences in decoding between areas were invariant to the algorithm used, still showing the dissimilarities found with HMM, thus indicating that these dissimilarities are intrinsic in the information encoded by parietal neurons. These results highlight that, when decoding from the parietal cortex, for example in brain-machine interface implementations, attention should be paid in selecting the most suitable source of neural signals, given the great heterogeneity of this cortical sector.

show abstract

Section: Resultsmentioning

confidence: 99%

Similar neural states, but dissimilar decoding patterns for motor control in parietal cortex

Vaccari,

Diomedi,

De Vitis

et al. 2024

Network Neuroscience

Self Cite

View full text Add to dashboard Cite

show abstract

“…This work presents the results of an extensive data curation effort, driven by the aspiration to enhance the accessibility of our research data. Our work is based on consolidating multiple datasets from various papers [8][9][10][11][12][13] into a single and easily accessible repository 14 , eliminating the need for researchers to request data. Within the manuscript, we detail the repository, providing essential insights for data reuse, with a clear and consistent terminology to describe the experimental task and the behavioral events recorded together with the neural activity.…”

Section: Background and Summarymentioning

confidence: 99%

“…The datasets here presented have been efficiently used in last works to functionally characterize individual neurons in the SPL, to model latent states and decode population activity 9,32 , but also to explain method of investigation 10,33 .…”

Section: Background and Summarymentioning

confidence: 99%

Neurophysiological recordings from parietal areas of macaque brain during an instructed-delay reaching task

Diomedi,

Vaccari,

Gamberini

et al. 2024

Sci Data

View full text Add to dashboard Cite

Facilitating data sharing in scientific research, especially in the domain of animal studies, holds immense value, particularly in mitigating distress and enhancing the efficiency of data collection. This study unveils a meticulously curated collection of neural activity data extracted from six electrophysiological datasets recorded from three parietal areas (V6A, PEc, PE) of two Macaca fascicularis during an instructed-delay foveated reaching task. This valuable resource is now accessible to the public, featuring spike timestamps, behavioural event timings and supplementary metadata, all presented alongside a comprehensive description of the encompassing structure. To enhance accessibility, data are stored as HDF5 files, a convenient format due to its flexible structure and the capability to attach diverse information to each hierarchical sub-level. To guarantee ready-to-use datasets, we also provide some MATLAB and Python code examples, enabling users to quickly familiarize themselves with the data structure.

show abstract

Using HMM to Model Neural Dynamics and Decode Useful Signals for Neuroprosthetic Control

Cited by 2 publications

References 53 publications

Similar neural states, but dissimilar decoding patterns for motor control in parietal cortex

Similar neural states, but dissimilar decoding patterns for motor control in parietal cortex

Neurophysiological recordings from parietal areas of macaque brain during an instructed-delay reaching task

Contact Info

Product

Resources

About