Reconstruction of micron resolution mouse brain surface from large-scale imaging dataset using resampling-based...

Area of Science:

• Neuroscience
• Computational Biology
• Medical Imaging

Background:

• Accurate brain surface profiling is crucial for neuroscience research, including registration, segmentation, and circuit mapping.
• High-throughput imaging now allows whole mouse brain visualization at micron resolution, necessitating advanced reconstruction techniques.
• Existing methods struggle with large datasets and inhomogeneous signal intensity, limiting detailed quantitative brain studies.

Purpose of the Study:

• To develop an efficient and accurate computational model for reconstructing micron-resolution brain surfaces from large-scale neuroimaging data.
• To overcome limitations of global analysis methods in handling large datasets and variable signal intensities.
• To provide a robust tool for detailed quantitative analysis in neuroscience.

Main Methods:

• A novel resampling-based variational model was proposed.
• The model fixes movement directions of initial boundary elements.
• Final boundary positions are determined by local signal intensity for precise surface reconstruction.

Main Results:

• The proposed method achieves high recall and precision rates exceeding 97%.
• It is significantly faster, hundreds-fold quicker than conventional methods like level set and active contour.
• Demonstrated fast reconstruction of whole mouse brain surfaces from hundreds of GB datasets within 6 hours.

Conclusions:

• The resampling-based variational model offers an effective solution for reconstructing detailed brain surfaces from large, complex neuroimaging datasets.
• This method significantly enhances the speed and accuracy of brain surface profiling.
• Enables more comprehensive quantitative studies in neuroscience by facilitating analysis of high-resolution brain data.