Nov 21 (Mon) @ 11:00am: "Segmentation, Tracking, and Shape Modeling for 3D Time-lapse Microscopy Images," Jiaxiang (Tom) Jiang, ECE PhD Defense

Abstract

3D Time lapse microscopy images are important in our understanding of structure and functions of cells.  With the growing amount of imaging data, quantitative analysis requires robust and scalable methods. Fundamental problems such as automatic segmentation of cells, subcellular feature extraction, tracking of such structures in 3D over time, and modeling 3D shapes in robust manner, remain. This dissertation focuses on two problems of broad interest: segmentation and tracking in 3D time-lapse confocal imaging data at the cellular level; and 3D shape models for classification of cells, specifically neuronal cells.

For the confocal imaging data, we develop a 3D analysis pipeline that works with membrane tagged data, to segment cells in 3D over time. In addition, we detect subcellular structures of interest, such as 3 cell wall junctions and lobes, that enable further quantification of the cell growth process. A graph-based tracking method is proposed. Detailed quantitative evaluation results demonstrate the robustness and state-of-the art performance of the proposed methods.

For 3D shape analysis, we propose a robust computational 3D skeleton model to analyze neuron morphology. It is the first deep learning method to compute a 3D neuron skeleton model directly from discrete 3D surface points for neuron classification. The main innovation is in formulating the learning problem associated with computing the medial axis transform that represents the 3D skeleton. Our method results in an accurate and robust skeleton representation and achieves state-of-the-art performance in classifying neuron types. The software and associated data are available via GitHub and as service through the BisQue infrastructure developed at UC Santa Barbara.

Bio

Jiaxiang (Tom) Jiang is a Ph.D. Candidate in the Department of Electrical and Computer Engineering at the University of California, Santa Barbara. His research interests span time-lapse 3D biomedical image analysis including segmentation, tracking, and shape modeling for microscopy images. He received his B.S. in Electronic Information Engineering from University of Electronic Science and Technology of China (UESTC) in 2016 and his M.S. in Electrical and Computer Engineering from The Ohio State University (OSU) in 2017.

Hosted by: Professor B.S. Manjunath

Submitted by: Jiaxiang (Tom) Jiang <jjiang00@ucsb.edu>