University of Twente Student Theses

Login

Tracking and Control of Soft, Self-Folding Miniaturized Agent using Ultrasound Images

Thirukokaranam Chandrasekar, Krishna Kumar (2016) Tracking and Control of Soft, Self-Folding Miniaturized Agent using Ultrasound Images.

[img]
Preview
PDF
3MB
Abstract:The usage of micro-sized agents in magnetic based drug delivery systems could be made a lot more affordable if it is possible to control these agents wirelessly based on feedback obtained from the ultrasound clinical images. However, since the image is of ultrasound type, the most common edge detection techniques used in digital image processing has to be modified and adapted in such a way, the control of the micro agent is made possible. The main reason is due to absence of sharp edges with the obtained image. The thesis mainly focuses on solving this problem. Also Control algorithms were written to control and drive the robot across various scenarios and motion profiles for performing pick and move operations. Thus the goal of this assignment is to track and control hydrogel micro-grippers using two-dimensional ultrasound images. In the initial step a tracker is designed which is evaluated with the collected datasets. The tracker is then implemented to detect the miniaturized gripper and to track it in the presence of occlusions. Following it, an experimental validation of the proposed technique is performed. After the evaluation, the proposed algorithm is used in the experimental setup to control the motion of the miniaturized agent along a defined path. Various motion experiments were performed in order to evaluate the performance of the proposed technique. Finally, with the help of proper motion profiles and control algorithms, the hydrogel gripper is made to reach a bead, grasp it and transport it to a specified target area. Also in order to evaluate clinically relevant scenarios, algorithms and advanced computer vision techniques were selected and compared for exploring the best possible features from the ultrasound images and eventually based on the results, a specialized tracker was formulated with aim of tracking the micro gripper in real tissue type of environment.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:33 physics, 35 chemistry, 42 biology, 51 materials science, 54 computer science
Programme:Human Media Interaction MSc (60030)
Link to this item:http://purl.utwente.nl/essays/71104
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page