2018-07-06 | 孔俊：Quantitative Analysis of Multi-Dimensional Big Histopathology Microscopy Image Data
In biomedical research, the availability of an increasing array of high-throughput and high-resolution instruments has given rise to large datasets of imaging data. These datasets provide highly detailed views of tissue structures at the cellular level and present a strong potential to revolutionize biomedical translational research. However, traditional human-based tissue review is not feasible to obtain this wealth of imaging information due to the overwhelming data scale and unacceptable inter- and intra-observer variability. In this talk, I will first describe how to efficiently process Two-Dimension (2D) digital microscopy images for highly discriminating phenotypic information with development of microscopy image analysis algorithms and Computer-Aided Diagnosis (CAD) systems for processing and managing massive in-situ micro-anatomical imaging features. Equipped with statistical machine learning techniques, these systems can automatically detect, measure, group, and classify a large scale of anatomical structures from microscopy images of histological specimens to support higher-level diagnosis and follow-up scientific investigations. Additionally, I will present novel algorithms to support Three-Dimension (3D), molecular, and time-lapse microscopy image analysis. Specifically, I will demonstrate an on-demand registration method within a dynamic multi-resolution transformation mapping and an iterative transformation propagation framework. This will allow us to efficiently scrutinize volumes of interest on-demand in a single 3D space. For segmentation, I will present a scalable segmentation framework for histopathological structures with two steps: 1) initialization with joint information drawn from spatial connectivity, edge map, and shape analysis, and 2) variational level-set based contour deformation with data-driven sparse shape priors. For 3D reconstruction, I will present a novel cross section association method leveraging Integer Programming, Markov chain based posterior probability modelling and Bayesian Maximum A Posteriori (MAP) estimation for 3D vessel reconstruction. I will also present new methods for multi-stain image registration, biomarker detection, and 3D spatial density estimation for For molecular imaging data integration. For time-lapse microscopy images, I will present a new 3D cell segmentation method with gradient partitioning and local structure enhancement by eigenvalue analysis with hessian matrix. A derived tracking method will be also presented that combines Bayesian filters with a sequential Monte Carlo method with joint use of location, velocity, 3D morphology features, and intensity profile signatures. Our proposed methods featuring by 2D, 3D, molecular, and time-lapse microscopy image analysis will facilitate researchers and clinicians to extract accurate histopathology features, integrate spatially mapped pathophysiological biomarkers, and model disease progression dynamics at high cellular resolution. Therefore, they are essential for improving clinical decisions, enhancing prognostic predictions, inspiring new research hypotheses, and realizing personalized medicine.
Dr. Jun Kong, is an Assistant Professor in the Department of Biomedical Informatics and Department of Mathematics and Computer Science at Emory University. He is also an adjunct faculty in Department of Biomedical Engineering in Georgia Institute of Technology. Furthermore, Dr. Kong is a member of the Cancer Cell Biology Program of the Winship Cancer Institute and directs a lab that develops biomedical image analysis algorithms, computer-aided diagnosis systems, large-scale integrative approaches, and high performance computing methods for quantitative cancer research. He is particularly interested in advancing translational oncology research with quantitative computer, mathematical, engineering, and informatics methods. He is committed to analyzing discriminating cancer pathologic hallmarks and signaling biomarkers highly correlated with clinical outcome, improving understanding about signature gene networks and cancer progression mechanisms, and defining effective molecular targets for personalized therapeutic medicine. He has developed a large number of image analysis systems and quantitative data integration methods for numerous cancer diseases, such as neuroblastoma, lymphoma, liver, and glioblastoma multiforme, with intensive development and use of image analysis and pattern recognition techniques for automated processing microscopy images of histological specimens. He has also deeply engaged in the In Silico integrative Brain Tumor Research in which he has developed automated analysis methods for whole-slide pathology image study, cohesive patient stratification with quantitative phenotypic imaging data, and discovery of inherent associations across phenotypic groups, molecular descriptions and clinical data. He has over 14 years of experience in cancer imaging data analysis, with nine years dedicated to quantitative brain tumor research. Dr. Kong has written over 80 peer-reviewed manuscripts in cancer research, computer, and biomedical informatics journals and top-notch conferences, with citations from researchers worldwide.