Abstract
Understanding the basis of brain function requires knowledge of cortical operations over wide spatial scales and the quantitative analysis of brain activity in well-defined brain regions. Matching an anatomical atlas to brain functional data requires substantial labor and expertise. Here, we developed an automated machine learning-based registration and segmentation approach for quantitative analysis of mouse mesoscale cortical images. A deep learning model identifies nine cortical landmarks using only a single raw fluorescent image. Another fully convolutional network was adapted to delimit brain boundaries. This anatomical alignment approach was extended by adding three functional alignment approaches that use sensory maps or spatial-temporal activity motifs. We present this methodology as MesoNet, a robust and user-friendly analysis pipeline using pre-trained models to segment brain regions as defined in the Allen Mouse Brain Atlas. This Python-based toolbox can also be combined with existing methods to facilitate high-throughput data analysis.
Highlights
Understanding the basis of brain function requires knowledge of cortical operations over wide spatial scales and the quantitative analysis of brain activity in well-defined brain regions
We suggest that functional maps that represent specific spatio-temporal consensus patterns of regional activation observed using activity sensors such as GCAMP62,30,31 or potentially hemodynamic activation[32,33] can be used for atlas registration
We present an open-source toolbox that will facilitate the analysis of mesoscale imaging data from wide-field microscopy
Summary
Understanding the basis of brain function requires knowledge of cortical operations over wide spatial scales and the quantitative analysis of brain activity in well-defined brain regions. This anatomical alignment approach was extended by adding three functional alignment approaches that use sensory maps or spatial-temporal activity motifs We present this methodology as MesoNet, a robust and user-friendly analysis pipeline using pre-trained models to segment brain regions as defined in the Allen Mouse Brain Atlas. For atlas-to-brain, we trained a deep learning model[25] to automatically identify cortical landmarks based on single raw fluorescence wide-field images. For the brain-to-atlas approach, our system automatically registers cortical images to a common atlas using predicted cortical landmarks This alignment approach, while robust in the presence of anatomical landmarks, does not leverage regional patterns within functional calcium imaging data that are related to underlying structural connectivity[2,30,31]. We demonstrate that this new open-source toolbox for automated brain image analysis is robust to morphological variation and can process multiple data sets in a relatively automated manner
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
