Robust Displacement Estimation for Ultrasound Elastography and Thermal Imaging
Embargo until
Date
2014-07-25
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Johns Hopkins University
Abstract
Ultrasound imaging is becoming the modality of choice for many diagnostic and surgical procedures. Besides being inexpensive and safe, ultrasonography is emerging as a quantitative tool able to image tissue properties. In this dissertation we focus on elastography and thermal imaging, which both rely on the measurement of real or apparent motion in ultrasound image sequences.
In ultrasound elastography, signal decorrelation is widely viewed as the major limiting factor for adoption of into clinical practice. In this dissertation we focus on improving the robustness of a displacement estimation method based on dynamic programming, addressing multiple weak points. We propose a set of tools which can improve its ability to overcome displacement discontinuities and regions of poorly correlated RF data. The method is further extended to three dimensional data. Phantom, animal and human studies are presented for experimental validation. The addition of robust tools results in an improved ability to achieve repeatable, artifact-free strain maps, without compromising computational speed.
In thermal imaging, we focus on the estimation of real and apparent motion while the tissue temperature is increased in an ablation procedure. Estimating heat-induced echo shifts is a very difficult problem because of their very small amplitude, on the order of tens of microns. They can easily be masked by other sources of deformation/movement from the environment such as patient motion or hand tremor. In this dissertation, we build upon the robust displacement estimation method for elastography, with the additional deployment of an iterative motion compensation algorithm. The validation experiments are performed on laboratory induced ablation lesions, where the ultrasound probe is either held by the operator's hand or supported by a robotic arm. We demonstrate the ability to detect and remove non-heat induced tissue motion at every step of the ablation procedure. Our results exceed the state of the art in both the accuracy of temperature estimation as well as the length of time over which temperature estimation can be performed. Previous research in the area of motion compensation resulted in good results for experiments lasting less than 10 seconds. Our experiments lasted close to 20 minutes.
Description
Keywords
ultrasound, elastography, thermal imaging, mosaicking, prostatectomy, liver ablation