2019年8月2日
MarmoNet: a pipeline for automated projection mapping of the common marmoset brain from whole-brain serial two-photon tomography
https://arxiv.org/abs/1908.00876
- 巻
- abs/1908.00876
- 号
- 記述言語
- 英語
- 掲載種別
- 機関テクニカルレポート,技術報告書,プレプリント等
Understanding the connectivity in the brain is an important prerequisite for<br />
understanding how the brain processes information. In the Brain/MINDS project,<br />
a connectivity study on marmoset brains uses two-photon microscopy fluorescence<br />
images of axonal projections to collect the neuron connectivity from defined<br />
brain regions at the mesoscopic scale. The processing of the images requires<br />
the detection and segmentation of the axonal tracer signal. The objective is to<br />
detect as much tracer signal as possible while not misclassifying other<br />
background structures as the signal. This can be challenging because of imaging<br />
noise, a cluttered image background, distortions or varying image contrast<br />
cause problems.<br />
We are developing MarmoNet, a pipeline that processes and analyzes tracer<br />
image data of the common marmoset brain. The pipeline incorporates<br />
state-of-the-art machine learning techniques based on artificial convolutional<br />
neural networks (CNN) and image registration techniques to extract and map all<br />
relevant information in a robust manner. The pipeline processes new images in a<br />
fully automated way.<br />
This report introduces the current state of the tracer signal analysis part<br />
of the pipeline.
understanding how the brain processes information. In the Brain/MINDS project,<br />
a connectivity study on marmoset brains uses two-photon microscopy fluorescence<br />
images of axonal projections to collect the neuron connectivity from defined<br />
brain regions at the mesoscopic scale. The processing of the images requires<br />
the detection and segmentation of the axonal tracer signal. The objective is to<br />
detect as much tracer signal as possible while not misclassifying other<br />
background structures as the signal. This can be challenging because of imaging<br />
noise, a cluttered image background, distortions or varying image contrast<br />
cause problems.<br />
We are developing MarmoNet, a pipeline that processes and analyzes tracer<br />
image data of the common marmoset brain. The pipeline incorporates<br />
state-of-the-art machine learning techniques based on artificial convolutional<br />
neural networks (CNN) and image registration techniques to extract and map all<br />
relevant information in a robust manner. The pipeline processes new images in a<br />
fully automated way.<br />
This report introduces the current state of the tracer signal analysis part<br />
of the pipeline.
- リンク情報
- ID情報
-
- DBLP ID : journals/corr/abs-1908-00876
- arXiv ID : arXiv:1908.00876