טכניון מכון טכנולוגי לישראל
הטכניון מכון טכנולוגי לישראל - בית הספר ללימודי מוסמכים  
M.Sc Thesis
M.Sc StudentNeustadter Hannah
SubjectNon-Rigid Registration Of Pancreas Subharmonic Ultrasound
Images
DepartmentDepartment of Biomedical Engineering
Supervisors Professor Emeritus Dan Adam
Professor Emeritus Eitan Kimmel


Abstract

Detection of abnormal vasculature in pancreatic cancer tumors may allow early diagnosis, more effective treatment and improvement of the survival rates of patients. Contrast enhanced ultrasound (CEUS) is a commonly used imaging modality that promotes visualization of vasculature of the tissue and may help detect small morphological changes. Non-rigid registration of images (frames) within a clip may be beneficial to overcome some of the problems inherent to the physiological motion of the observed anatomical structures. Herein we present an improved visibility of the pancreas vasculature by performing non-rigid registration of the images acquired with the fundamental harmonic imaging (FHI) and applying the estimated deformation field to the images obtained by employment of subharmnic imaging (SHI).  

Two groups of data were used: 1- human pancreas with pancreatic tumor and 2 - mice hidlimb with induced tumor. The data were acquired using a curve-linear transducer (transmissions of 4-MHz for the fundamental frequency called B-mode and 2.5-MHz for the subharmonic frequency) and GE Healthcare research LOGIQ E9 system. The  FHI and SHI were acquired in an interleaving acquisition mode, and are shown simultaneously. The clips were acquired after injecting a bolus of ultrasound contrast agent.

The Morphon (i.e. a local phase) registration method was used, by applying directional quadrature filters to the images. The registration produces a deformation field, which represents the new position of each pixel in the deformed image, so that it best matches the reference image. This deformation field is then applied to the SHI clip.

The results, demonstrate an increase of the correlation among the images. In group 1, first a significant improvement of the correlation is depicted after performing registration at the end-expiration period. Further improvement is achieved for the registration of a full breathing cycle. In the MIP images of the unprocessed FHI and SHI data, the edges of the vessel are smudged due to its movement, as opposed to the MIP result after applying the registration data.

In Conclusion, Both the clinical and the pre-clinical data, were acquired under conditions that severly limit the ability to register and process the images such as very low frame-rates.  Moreover, the FHI and SHI are acquired in an interleaving mode, leading to temporal mismatch due to the time gap between acquisitions. The low frame-rates affect the quality of motion estimation when movements occur, and the mismatch between the two clips does not allow proper correction. In addition, all the FHI clips and especially the pancreas, suffer from a low SNR caused by a significant clutter noise which interferes with the pre-processing steps and registration.

The correlation between the images of the SHI clips was improved for both groups, indicating an improved allignment. The visualization of the tissue vasculature has also been improved after the registration. The human pancreas data, which commonly suffers from large movements and low FPS, benefits from the two-step registration. When only small movments are present, there is no need for the two stages’ solution. Therefore, we can see good registration results in large movments clip such as the human pancreas.