Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
April 30, 2026Machine Learning with ApplicationsOpen Access

Dissecting the Computational Necessity of Hybrid Deep Learning Models for Decoding Complex Motor Behavior from Neuronal Population Activity

View Full Paper
Ask AI
Bookmark
Share

Authors

GMGhazal MirzaeeRLRoshanak LatifikhereshkiJCJonathan Chang

Discussion

Loading...

Member takes

Overview

Randomized trial reveals effective decoding of motor behavior in neuronal populations, indicating the importance of hybrid models.

Key Points

  • This research aims to improve the decoding of complex motor behaviors from neuronal population activity.
  • Employed an attention-based hybrid CNN-BiLSTM architecture to decode skilled forelimb movements.
  • Analyzed neuronal population activity using in vivo two-photon calcium imaging.
  • Conducted ablation analyses to explore the importance of spatial and temporal modeling.
  • Achieved reliable decoding of ipsilateral and contralateral forelimb movements during skilled locomotion.
  • Demonstrated that both spatial organization and temporal structure are crucial for decoding performance.

Cite This Study

Mirzaee et al. (2026) studied this question.

synapsesocial.com/papers/69f2f0e31e5f7920c6386e7ahttps://doi.org/10.1016/j.mlwa.2026.100905
View Full Paper
Ask AI
Bookmark
Share