Practical Approach for Automation of High Fidelity Haptic Models Generation
thesisposted on 18.02.2018, 00:00 by Mina Haratiannezhadi
This work describes the process for turning a raw volume into a 3-Dimensional volume that can provide haptic feedback and suggests a novel approach for automation of this process. The generated files are runnable in ImmersiveTouch simulators and present the properties of a haptic virtual object. ImmersiveTouch simulators perform various surgical procedures and they have been used in many hospitals and medical schools for educational purposes. Since these simulators are capable of performing unlimited customized cases, it is not efficient to employ a technician in each institution to generate the models manually. Existing approaches need the user to know the framework of an object to apply the required changes to the template file. Making these objects requires a thorough knowledge of the simulator dataset format and structure of files needed to run the model. One of the fundamental problems in this area is to find a solution to simplify this process for nonprogrammer users to customize virtual models. In this research, three softwares were designed and implemented to automate three main steps of object generation. The purpose of this practice is to automate all of the process of making the haptic model. The main input of this process is medical images, which are in different sizes, and formats. All of the images should be converted to the standard 8bit format, and the noises should be eliminated. Different tissues must be segmented separately; in this case a mask might be needed to apply on an image to take a specific tissue. After all the required input files are prepared, the user inserts haptic properties of the object into the application, and the application generates the required files by viii extracting header information of the image. ImmersiveTouch simulators could run the output of the software and display the virtual objects.