Funded ProjectAutomatic Segmentation of single cell anatomy
Analysis of three-dimensional biological cell samples is critical for understanding the mechanisms of disease and for the development of specific treatment of disorders. Anatomy of single cells can be used to assess cell cycle, distinguish between benign and malignant tumor cells, or tell about infection pathway, ultimately helping in the development of specific drugs. Soft x-ray microscopy is the unique technology that can image whole intact cells under normal and pathological conditions without labelling or fixation, at high throughput and spatial resolution. Ongoing improvements in single cell imaging result in faster acquisition times, hence increasing the demand for accelerated image analysis. At present, image segmentation is a major bottleneck in the soft x-ray microscopy data analysis pipeline. Currently, segmentation is a primarily manual task, which is tedious and time consuming. Automatic segmentation is a difficult task because the low contrast to noise ratio makes conventional unsupervised techniques ineffective, and the limited availability of already-labeled data makes it difficult to use supervised techniques. We aim to build and deliver open-source software to address automatic segmentation of single cell anatomy. Our free and easy-to-use open-source online platform (Biomedisa.org) developed for semi-automatic segmentation of large volumetric images (organs and organisms) is based on a smart interpolation of sparsely pre-segmented slices. We aim to extend this semi-automatic segmentation platform for single cell tomography. For validation of segmentation pipeline, we will use soft x-ray microscopy data of uninfected and infected human lung epithelial cells, thus directly helping to identify individual factors and possible targets for broad-spectrum antivirals.
