Eng    中   
 
  首頁 > 活動

2004 一月   二月   三月   四月   五月   八月   九月   十月   十一月   十二月  
2005 一月   二月   三月   四月   五月   六月   七月   八月   九月   十月   十一月   十二月  
2006 一月   二月   三月   四月   五月   六月   七月   九月   十月   十一月   十二月  
2007 一月   二月   三月   四月   五月   六月   七月   八月   九月   十月   十一月   十二月  
2008 一月   二月   三月   四月   五月   六月   七月   八月   九月   十月   十一月   十二月  
2009 一月   二月   三月   四月   五月   六月   七月   八月   九月   十月   十一月   十二月  
2010 一月   二月   三月   四月   五月   六月   九月   十月   十一月   十二月  
2011 一月   二月   三月   四月   五月   六月   七月   八月   九月   十月   十一月   十二月  
2012 一月   二月   四月   五月   六月   七月   八月   九月   十一月   十二月  
2013 一月   二月   三月   四月   五月   六月  

Event(s) on September 2007


  • 11/9/2007

    題目: Global Conformal Parametrization and its Applications to Human Brain Mapping Research
    講員: Mr. Ronald Lok Ming Lui, Department of Mathematics, UCLA, USA
    時間/地點: 11:30  -  12:30
    RRS 905
    摘要: Human brain cortical surface is a very complicated surface. Computational problems arise when integrating and comparing brain data. In order to analyze and compare brain data more effectively, it is desirable to map the surface to some simple domain, such as the sphere or rectangles, while retaining the geometric information on the surface. Conformal mapping is one type of maps that effectively fits this purpose. In this talk, I will describe three effective algorithms for conformal parametrization of the brain surface. It allows us to map the brain surfaces to the sphere, rectangle or punctured disk. With these techniques, a lot of human brain mapping problems can be simplified. In the second part of my talk, I will describe how we can apply the conformal parametrization to various medical applications. In particular, I will describe how we can use the conformal map to automatic detection of brain anatomical features, solving variational problems on surfaces, landmark matching registration between brain surfaces, etc.


  • 18/9/2007

    題目: Data Envelopment Analysis and its Applications
    講員: Prof. Cecilio Mar Molinero, Kent Business School, University of Kent, UK
    時間/地點: 11:30  -  12:30
    FSC 1217
    摘要: Data Envelopment Analysis (DEA) is now well established as a tool for permance assessment. DEA relies on a series of multiple comparisons between units that use the same inputs to produce the same outputs, and determines which units make best use of the available resources. DEA is not a statistical technique and it is difficult to establish if an input or an output is necessary in the formulation or it contributes little to the model. In this seminar, I will present an approach to the analysis of DEA specification based on the estimation of a multiplicity of models, whose similarities and differences are later explored by means of multivariate analysis techniques. I will present examples of application of the technique to the exploration of microcredit institutions (banking institutions that make small loans to the poor), and to the analysis of risks and rewards in a portofolio of insurance policies held by a major Spanish company.

Copyright © 2012 Hong Kong Baptist University. All rights reserved.
HKBU Math Dept HKBU