A 3D registration method for US/PET multi-modality small animal tumor imaging
Journal
Proceedings - IEEE Ultrasonics Symposium
Journal Volume
1
Pages
2214-2217
Date Issued
2006
Author(s)
Abstract
Small animal models have been used extensively in disease research, genomics research, drug development and developmental biology. Non-invasive, small animal imaging with adequate spatial resolution and sensitivity is therefore beneficial to such research fields. In particular, multi-modality small animal imaging can provide complementary image information and is therefore of great interest. In this paper, a method for registering high frequency ultrasonic images (microUS) with microPET images is presented. A registration phantom with six multi-modality image markers is custom-designed and placed in the small animal holder. The multi-modality image marker is made of a glass bead with 0.43-0.60 mm diameter and 1 μl [18F]FDG is placed in each marker position. The small animal holder is used to transfer mice between the microPET (R4, Concord Microsystems, Knoxville, TN) and the microUS systems (CLI 1500Ti, Capistrano Labs, San Clemente, CA). Multi-modality imaging is performed on C57BL/6J black mice bearing WF-3 ovary cancer cells from the second week after tumor implantation. Rigid body image registration of the six markers is performed. The average registration error is 0.37 mm when all six markers are used. Also, the target registration error increases when the number of markers decreases. After image registration, image segmentation and fusion is then performed for the tumor. With the multi-modality small animal imaging method, the structural information from microUS can be combined with the functional information from microPET. The proposed method has been shown to be an effective tool for cancer research. ? 2006 IEEE.
SDGs
Other Subjects
3D registration; Cancer research; Developmental biology; Disease research; Drug development; Effective tool; Functional information; Genomics; Glass bead; High frequency; Image information; MicroPET; Multi-modality; Multi-modality images; Non-invasive; Ovary cancer; Registration error; Research fields; Small Animal; Small animal imaging; Small animal model; Spatial resolution; Structural information; Target registration errors; Tumor imaging; Ultrasonic images; Animals; Image fusion; Image registration; Research; Tumors; Ultrasonic imaging; Ultrasonics; Image segmentation
Type
conference paper