Current Research

3D Optical-CT Single Cell Image Analysis
Yiting Xie

Three-dimensional single cell analysis is performed using optical CT microscope images. In the imaging process, a single cell is suspended in an optical gel and injected into a tube. Images from multiple perspectives are taken as the tube rotates and are reconstructed into a 3D image using the filtered back-projection method [1-3]. Compared to 2D cell images, 3D images provide more information and are not affected by cell orientations [1, 3].

Our database consists of 248 scans from 5 cell types: AF cancer cells, Columnar, Macrophage, Metaplastic and Squamous (see Fig.1). Cell nucleus and cytoplasm regions are segmented using two fully automated methods: a gradient-based method and a sequential graph-cut method. The sequential graph-cut method had a superior performance than the gradient method when compared to manual reference boundaries (Dice Similarity Coefficient 0.86 for nucleus and 0.72 for cytoplasm using graph-cut) [4].

Figure 1. Five types of cells. Upper row shows an axial CT slice and lower row shows the 3D visualization of segmented nucleus (brown) and cytoplasm (green).

Publications

1. A. P. Reeves, E. J. Seibel, M. G. Meyer, T. Apanasovich, and A. M. Biancardi. Nuclear cytoplasmic cell evaluation from 3D Optical CT Microscope Images. SPIE Medical Imaging, Feb 2012.
2. Q. Miao, A. P. Reeves, F. W. Patten, and E. J. Seibel. Multimodal 3D Imaging of Cells and Tissue, Bridging the Gap Between Clinical and Research Microscopy. Annals of Biomedical Engineering, 2011.
3. N. Agarwal, A. M. Biancardi, F. W. Patten, A. P. Reeves and E. J. Seibel. Quantification of Relative Chromatin Content in Flow Cytometry Standards using 3D OPTM Imaging Technique. SPIE Medical Imaging, Feb 2013.
4. Y. Xie and A. P. Reeves. Single 3D cell segmentation from optical CT microscope images. SPIE Medical Imaging, Feb 2014.
   List of Current Research Projects