NEATmap: a high-efficiency deep learning approach for whole mouse brain neuronal activity trace mapping.

Area of Science:

• Neuroscience
• Computational Biology
• Bioimaging

Background:

• Quantitative analysis of neuronal activation is crucial for understanding brain function.
• Analyzing whole-brain neuronal activity in large mouse cohorts from terabytes of imaging data presents significant challenges in consistency and comprehensiveness.
• Existing methods struggle with the scale and complexity of volumetric imaging data for neuronal activity tracing.

Purpose of the Study:

• To introduce NEATmap, a novel deep learning-based software for high-efficiency and high-precision whole-brain neuronal activity mapping.
• To enable automated segmentation and quantitative analysis of immunofluorescence-labeled c-Fos+ neurons.
• To facilitate the study of brain-wide differentiated neuronal activation patterns in response to various stressors.

Main Methods:

• Development of NEATmap, a user-friendly software utilizing deep learning algorithms.
• Automated segmentation of immunofluorescence-labeled c-Fos+ neurons within whole mouse brains.
• Quantitative analysis of neuronal activation traces from large volumetric imaging datasets.

Main Results:

• NEATmap demonstrates high efficiency and precision in mapping neuronal activity across the entire mouse brain.
• The software successfully automated the segmentation and quantification of c-Fos+ neurons.
• Application of NEATmap revealed differentiated neuronal activation patterns in response to physical and psychological stressors.

Conclusions:

• NEATmap provides a robust solution for the comprehensive analysis of whole-brain neuronal activity.
• The software significantly overcomes the challenges associated with large-scale volumetric imaging data analysis.
• NEATmap is a valuable tool for neuroscience research, particularly in understanding stress-related neural circuits.