CNRS, UMR 5292, INSERM, U1028, Lyon Neuroscience Research Center, Neuroplasticity and Neuropathology of Olfactory Perception Team, Lyon, F-69000, France; University Lyon, F-69000, France; University Lyon 1, Villeurbanne, F-69000, France.
CNRS, UMR 5292, INSERM, U1028, Lyon Neuroscience Research Center, Neuroplasticity and Neuropathology of Olfactory Perception Team, Lyon, F-69000, France; University Lyon, F-69000, France; University Lyon 1, Villeurbanne, F-69000, France.
J Neurosci Methods. 2018 Jul 1;304:136-145. doi: 10.1016/j.jneumeth.2018.04.004. Epub 2018 Apr 22.
Cellular imagery using histology sections is one of the most common techniques used in Neuroscience. However, this inescapable technique has severe limitations due to the need to delineate regions of interest on each brain, which is time consuming and variable across experimenters.
We developed algorithms based on a vectors field elastic registration allowing fast, automatic realignment of experimental brain sections and associated labeling in a brain atlas with high accuracy and in a streamlined way. Thereby, brain areas of interest can be finely identified without outlining them and different experimental groups can be easily analyzed using conventional tools. This method directly readjusts labeling in the brain atlas without any intermediate manipulation of images.
We mapped the expression of cFos, in the mouse brain (C57Bl/6J) after olfactory stimulation or a non-stimulated control condition and found an increased density of cFos-positive cells in the primary olfactory cortex but not in non-olfactory areas of the odor-stimulated animals compared to the controls.
COMPARISON WITH EXISTING METHOD(S): Existing methods of matching are based on image registration which often requires expensive material (two-photon tomography mapping or imaging with iDISCO) or are less accurate since they are based on mutual information contained in the images. Our new method is non-imaged based and relies only on the positions of detected labeling and the external contours of sections.
We thus provide a new method that permits automated matching of histology sections of experimental brains with a brain reference atlas.
使用组织学切片进行细胞成像，是神经科学中最常用的技术之一。然而，这种不可避免的技术存在严重的局限性，因为需要对每个大脑中的感兴趣区域进行描绘，这既耗时又因实验者而异。
我们开发了基于向量场弹性配准的算法，允许快速、自动地对实验脑切片进行重新定位，并以高精度和简化的方式在脑图谱中对相关标签进行关联。这样，无需描绘感兴趣区域，就可以精细地识别脑区，并且可以使用常规工具轻松分析不同的实验组。该方法直接在脑图谱中重新调整标签，而无需对图像进行任何中间操作。
我们在嗅觉刺激或未刺激对照条件下对小鼠大脑（C57Bl/6J）中的 cFos 表达进行了映射，发现与对照相比，在气味刺激动物的初级嗅觉皮层中，cFos 阳性细胞的密度增加，但在非嗅觉区域中没有增加。
现有的匹配方法基于图像配准，这通常需要昂贵的材料（双光子断层扫描映射或 iDISCO 成像），或者不太准确，因为它们基于图像中包含的互信息。我们的新方法是非成像的，仅依赖于检测到的标签位置和切片的外部轮廓。
因此，我们提供了一种新方法，允许将实验大脑的组织学切片与大脑参考图谱自动匹配。
