Gal Mishne

This paper improves uncertainty estimation in brain-to-text BCIs by modifying training objectives and evaluation metrics, leading to more reliable and informative predictions.

This paper introduces a compositional decoding model for handwriting BCIs that improves generalization by leveraging behavioral motifs.

The paper identifies cell-type-specific activity dynamics in striatal neurons during movement by analyzing cortical and thalamic inputs in mice.

The paper improves the GraFT algorithm for functional calcium imaging by enhancing speed and accessibility through efficient solvers, compressive sensing, and a graphical user interface.

This paper introduces a novel dimensionality reduction technique that generates low-dimensional embeddings while preserving the global structure within the data for a variety of biological and non-biological datasets.

The paper shows that hemodynamic responses in the cortex are influenced by neuromodulators like norepinephrine, with distinct impulse response functions for different neuromodulatory signals.

GraFT isolates fluorescing components in neural imaging by learning time traces with spatial maps as coefficients, using a graph filtering model based on temporal activity rather than spatial proximity.

This study uses calcium imaging in awake mice to show that rapid changes in cortical functional connectivity encode spontaneous behaviors and reveal subnetworks not captured by traditional parcellation.

This paper uses non-linear manifold learning on fMRI data to uncover a shared low-dimensional space of brain dynamics across different cognitive tasks.

The paper introduces a method for extracting cellular structures and denoising temporal signals in calcium imaging data using graph-based clustering and nonlinear mapping.