High-Precision, Vision-Based Human Body Tracking


Helical Tomotherapy Device

This is my Ph.D. thesis work. The motivating problem is based around the current limitations of the helical tomotherapy device. The helical tomotherapy device is a method to deliver high radiation dosage to certain body areas, while leaving other areas with lower dosage. This type of radiation therapy is known as conformal treatment, as the radiation delivered conforms to the shape of the target, sparing other areas of the body high radiation exposure. A primary limitation of the helical tomotherapy device is its inability to sense patient movement, making it impossible to adjust the radiation delivery plan to compensate.

The goal of my thesis is to develop a minimally invasive, full-body, patient tracking system to be used with the helical tomotherapy system. To reliably treat patients, it is vitally important to detect when a patient becomes misaligned during treatment and adapt to it. The tolerances are very tight: movements of just five millimeters out of alignment can adversely affect treatment. At this point, detection of any movement would be an improvement as the device currently has no feedback mechanism to report on the patient's position while treatment is underway. Adding visual feedback to the helical tomotherapy device should improve the safety of the patient, reduce the considerable treatment time, and improve the cure rate of treatment.


  • Nathaniel Bird, "Calibration and Component Placement in Structured Light Systems for 3D Reconstruction Tasks," Ph.D. Dissertation, University of Minnesota, October 2009.
  • Nathaniel Bird and Nikolaos Papanikolopoulos, "Optimal Image-Based Euclidean Calibration of Structured Light Systems in General Scenes," IEEE Transactions on Automation Science and Engineering, volume 8, number 4, pp. 815--823, 2011.
  • Nathaniel Bird and Nikolaos Papanikolopoulos, "Placement Quality in Structured Light Systems," IEEE/RSJ Conference on Intelligent Robots and Systems, October 2009.