GraFT: Graph Filtered Temporal Dictionary Learning for Functional Neural Imaging DOI Creative Commons
Adam S. Charles, Nathan Cermak, Rifqi O. Affan

et al.

IEEE Transactions on Image Processing, Journal Year: 2022, Volume and Issue: 31, P. 3509 - 3524

Published: Jan. 1, 2022

Optical imaging of calcium signals in the brain has enabled researchers to observe activity hundreds-to-thousands individual neurons simultaneously. Current methods predominantly use morphological information, typically focusing on expected shapes cell bodies, better identify field-of-view. The explicit shape constraints limit applicability automated identification other important scales with more complex morphologies, e.g., dendritic or widefield imaging. Specifically, fluorescing components may be broken up, incompletely found, merged ways that do not accurately describe underlying neural activity. Here we present Graph Filtered Temporal Dictionary (GraFT), a new approach frames problem isolating independent as dictionary learning problem. focus time-traces-the main quantity used scientific discovery-and learn time trace spatial maps acting presence coefficients encoding which pixels time-traces are active in. Furthermore, novel graph filtering model redefines connectivity between terms their shared temporal activity, rather than proximity. This greatly eases ability our method handle data non-local structure. We demonstrate properties method, such robustness morphology, simultaneously detecting different neuronal types, and implicitly inferring number neurons, both synthetic real examples. applications at dendritic, somatic, scales.

Language: Английский

Distributed and Localized Dynamics Emerge in the Mouse Neocortex during Reach-to-Grasp Behavior DOI Creative Commons

Eros Quarta,

Alessandro Scaglione,

Jessica Lucchesi

et al.

Journal of Neuroscience, Journal Year: 2021, Volume and Issue: 42(5), P. 777 - 788

Published: Nov. 3, 2021

A long-standing question in systems neuroscience is to what extent task-relevant features of neocortical processing are localized or distributed. Coordinated activity across the neocortex has been recently shown drive complex behavior mouse, while selected areas canonically associated with specific functions (e.g., movements case motor cortex). Reach-to-grasp (RtG) known be dependent on circuits neocortex; however, global during these largely unexplored mouse. Here, we characterized, using wide-field calcium imaging, neocortex-wide dynamics mice either sex engaging an RtG task. We demonstrate that, beyond regions, several areas, such as visual and retrosplenial cortices, also increase their levels successful RtGs, homologous regions ipsilateral hemisphere involved. Functional connectivity among increases transiently around movement onset decreases movement. Despite this phenomenon, neural correlate kinematics measures RtGs sensorimotor only. Our findings establish that distributed co-orchestrate efficient control movements. SIGNIFICANCE STATEMENT Mammals rely reaching grasping for fine-scale interactions physical world. In cortex critical execution behavior, yet little about patterns areas. Using mesoscale-level networks a model cortical processing, investigated hypothesis could participate planning execution, indeed large network involved performing RtGs. Movement correlates mostly By demonstrating fine coexist mouse RtG, offer unprecedented view mammalian control.

Language: Английский

Citations

24
Irregular optogenetic stimulation waveforms can induce naturalistic patterns of hippocampal spectral activity DOI Creative Commons
Eric R. Cole, Thomas E. Eggers, David Weiß

et al.

Journal of Neural Engineering, Journal Year: 2024, Volume and Issue: 21(3), P. 036039 - 036039

Published: June 1, 2024

. Therapeutic brain stimulation is conventionally delivered using constant-frequency pulses. Several recent clinical studies have explored how unconventional and irregular temporal patterns could enable better therapy. However, it challenging to understand which are most effective for different therapeutic applications given the massively high-dimensional parameter space.

Language: Английский

Citations

3
Nonnegative matrix factorization for analyzing state dependent neuronal network dynamics in calcium recordings DOI Creative Commons
Daniel Carbonero, Jad Noueihed, Mark Kramer

et al.

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Nov. 13, 2024

Calcium imaging allows recording from hundreds of neurons in vivo with the ability to resolve single cell activity. Evaluating and analyzing neuronal responses, while also considering all dimensions data set make specific conclusions, is extremely difficult. Often, descriptive statistics are used analyze these forms data. These analyses, however, remove variance by averaging responses across sessions, or combinations neurons, create quantitative metrics, losing temporal dynamics activity, their relative each other. Dimensionally Reduction (DR) methods serve as a good foundation for analyses because they reduce into components, still maintaining variance. Nonnegative Matrix Factorization (NMF) an especially promising DR analysis method activity recorded calcium its mathematical constraints, which include positivity linearity. We adapt NMF our compare performance alternative dimensionality reduction on both artificial find that well-suited recordings, accurately capturing underlying data, outperforming common use.

Language: Английский

Citations

3
Statistical neuroscience in the single trial limit DOI
Alex H. Williams, Scott W. Linderman

Current Opinion in Neurobiology, Journal Year: 2021, Volume and Issue: 70, P. 193 - 205

Published: Oct. 1, 2021

Language: Английский

Citations

18
GraFT: Graph Filtered Temporal Dictionary Learning for Functional Neural Imaging DOI Creative Commons
Adam S. Charles, Nathan Cermak, Rifqi O. Affan

et al.

IEEE Transactions on Image Processing, Journal Year: 2022, Volume and Issue: 31, P. 3509 - 3524

Published: Jan. 1, 2022

Optical imaging of calcium signals in the brain has enabled researchers to observe activity hundreds-to-thousands individual neurons simultaneously. Current methods predominantly use morphological information, typically focusing on expected shapes cell bodies, better identify field-of-view. The explicit shape constraints limit applicability automated identification other important scales with more complex morphologies, e.g., dendritic or widefield imaging. Specifically, fluorescing components may be broken up, incompletely found, merged ways that do not accurately describe underlying neural activity. Here we present Graph Filtered Temporal Dictionary (GraFT), a new approach frames problem isolating independent as dictionary learning problem. focus time-traces-the main quantity used scientific discovery-and learn time trace spatial maps acting presence coefficients encoding which pixels time-traces are active in. Furthermore, novel graph filtering model redefines connectivity between terms their shared temporal activity, rather than proximity. This greatly eases ability our method handle data non-local structure. We demonstrate properties method, such robustness morphology, simultaneously detecting different neuronal types, and implicitly inferring number neurons, both synthetic real examples. applications at dendritic, somatic, scales.

Language: Английский

Citations

13