Segmentation of High-Resolution Human Brain Cryosections

We developed a semi-automatic technique for segmenting a large cryo-sliced human brain data set that contains 753 high-resolution RGB color images. This human brain data set presents a number of unique challenges to segmentation and visualization due to its size (over 7 GB) as well as the fact that each image not only shows the current slice of the brain but also unsliced "deeper layers" of the brain. These challenges are not present in traditional MRI and CT data sets. We have found that segmenting this data set can be made easier by using YIQ color model and morphology.We have used a hardware-assisted interactive volume renderer to evaluate our segmentation results. The segmentation is performed by two steps. First, we explore the color cues by converting the RGB encoded color information to the YIQ color model. Second, a filtering pipeline is applied using YIQ thresholding, median filter, region size thresholding, morphological operations, and RGB thresholding. The pictures below show the single steps of the filtering pipeline from an original slice to the segmented slice. Finally, we use volume rendering for three-dimensional visualization of the whole segmented data set. Reference: Ikuko Takanashi, Eric Lum, Kwan-Liu Ma, Joerg Meyer, Bernd Hamann, Arthur J. Olson, Segmentation and 3D Visualization of High- Resolution Human Brain Cryosections, Proceedings of Visualization and Data Analysis 2002, part of IS&T/SPIE's Conference, January 2002.