| Description |
Magnetic resonance imaging (MRI) is a medical imaging technique used in radiology to form pictures of the anatomy and the physiological processes of the body in both health and disease. MRI scanners use strong magnetic fields, radio waves, and field gradients to generate images of the organs in the body, including the brain (Wikipedia, https://en.wikipedia.org/wiki/Magnetic_resonance_imaging).
In order to investigate the organization of the brain and the mechanisms of its blood supply, it is imperative to combine data that covers the entire brain (such as MRI of the brain) with high-resolution optical microscopy images of processed tissue (such as histology and immunohistochemistry images).
The overall aim of the project is to quantify microstructural features at high-resolution in 3D. These include:
- Local density of excitatory and inhibitory neurons,
- Local distribution of neuron sizes,
- The local density of axons and dendrites
The specific aims include:
1. Develop a deep learning algorithm for tissue classification: segmenting the cerebral cortex in 2D histology images and separately in structural MRI. This will constrain and facilitate subsequent alignment of digitized histology images to the corresponding MRI data.
2. Develop pipelines for aligning 2D histology images to high-resolution ex-vivo MRI of small blocks of tissue from which the histology data were obtained.
3. Develop pipelines for aligning high-resolution MRI of small blocks of tissue to whole brain ex-vivo MRI data.
4. Develop image processing deep learning-based algorithm for identifying excitatory and inhibitory neurons, axons and dendrites in high-resolution histology and immunohistochemistry images.
5. Develop image processing deep learning-based algorithm for identifying blood vessels of the brain and reconstruct a high-resolution 3D volume of the blood vessels by combining MRI and optical microscopy data.
Techniques that will be pursued include:
a. Brain specific design and 3D-printing of a mold for coarse alignment of histology data to the corresponding ex-vivo MRI (see our paper, *Boopathy Jegathambal S, *Mok K, Rudko DA, Shmuel A (2018) MRI based brain-specific 3D-printed model aligned to stereotactic space for registering histology to MRI. Proceedings of the 40th conference of IEEE International Engineering in Medicine and Biology Society, Honolulu, Hawaii, July 2018: 802-805. doi: 10.1109/EMBC.2018.8512346.
b. Image processing, for registration/alignment of the different data modalities, and for identifying microstructural constituents of the brain, including excitatory and inhibitory neurons, axons, dendrites, blood vessels and capillaries.
c. Deep learning, for identifying microstructural constituents of the brain, including excitatory and inhibitory neurons, axons, dendrites, blood vessels and capillaries.
d. Software design and implementation, for creating pipelines by using existing scripts from NIH-cores available software packages and scripts.
|