Journal
MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING
Volume 60, Issue 2, Pages 487-500Publisher
SPRINGER HEIDELBERG
DOI: 10.1007/s11517-021-02495-8
Keywords
Image representation; Image registration; Neural networks; Regional localization; Neuron count
Ask authors/readers for more resources
This paper proposes a new method that uses neural networks to register different modal mouse brain slices and brain atlas, and completes the regional localization of the brain slices and counting of labeled neurons.
An important step in brain image analysis is to divide specific brain regions by matching brain slices to standard brain reference atlases, and perform statistical analysis on the labeled neurons in each brain region. Taking mouse fluorescently labeled brain slices as an example, due to the noise and distortion introduced during the preparation of brain slices, and the modal differences with standard brain atlas, the brain slices cannot directly establish an accurate one-to-one correspondence with the brain atlas, which in turn affects the accuracy of the number of labeled neurons in each brain region. This paper introduces the idea of image representation, uses neural networks to realize the registration of different modal mouse brain slices and brain atlas, completes the regional localization of the brain slices, and uses threshold segmentation to detect and count the labeled neurons in each brain region. The method proposed in this paper can effectively solve the problem of large deviation of neurons count caused by the inaccurate division of brain regions in large deformed brain slices, and can automatically realize accurate count of labeled neurons in each brain region of brain slices.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available