Queen's School of Computing

November 18th at10:30AM, Dupuis 217 


Registration problems requirefinding the mapping of information acquired in one coordinate system to relatedinformation acquired in a second (possibly the same as the first) coordinatesystem. In many computer-assisted surgical applications, the patient's anatomymust be registered to a pre-operative plan before navigational information canbe conveyed to the surgeon; thus, registration is a single-point of failure insuch applications. It is important that designers and users ofcomputer-assisted surgical systems understand how measurement errors that occurduring the registration process affect the accuracy of the computed registrationtransformation. In the first part of this talk I will describe the registrationproblem and define the basic terminology used to describe registration error incomputer-assisted surgical applications. I will then show how we can constructa model of registration error that we can use to predict its behaviour. In thesecond part of this talk I will discuss how the topic of registration errorcould be incorporated into an undergraduate and/or graduate course.


Burton Ma is anAssociate Professor in the Department of Computer Science and ElectricalEngineering at York University where he has been on faculty since 2009. Hecompleted his Ph.D. and undergraduate studies at Queen's University. Hisresearch work focuses on registration and calibration problems incomputer-assisted surgical applications. He has received multiple teachingawards at York University as well as the Howard Stavely Award for Excellence inTeaching from the School of Computing at Queen's University.