Subject-specific real-time respiratory liver motion compensation method for ultrasound-MRI/CT fusion imaging

• Published in: International journal for computer assisted radiology and surgery Vol. 10; no. 5; pp. 517 - 529
• Main Authors: Yang, Minglei, Ding, Hui, Kang, Jingang, Zhu, Lei, Wang, Guangzhi
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.05.2015
• Subjects: Computer Imaging; Computer Science; Health Informatics; Humans; Imaging; Liver - diagnostic imaging; Liver - pathology; Magnetic Resonance Imaging - methods; Medicine; Medicine & Public Health; Motion; Multimodal Imaging - methods; Original Article; Pattern Recognition and Graphics; Radiology; Respiration; Surgery; Tomography, X-Ray Computed - methods; Ultrasonography - methods; Vision; Fusion errors; Real-time motion compensation; Ultrasound-MRI/CT fusion imaging; Respiratory liver motion
• ISSN: 1861-6410; 1861-6429; 1861-6429
• DOI: 10.1007/s11548-014-1085-x

Purpose Ultrasound-MRI/CT fusion imaging is widely used in minimal invasive surgeries, such as liver biopsy and tumor ablation. However, respiration-induced quasi-periodic liver motion and deformation cause unacceptable misalignment of the fusion images (i.e., fusion error). A subject-specific liver motion model based on skin-mounted position sensor and corresponding ultrasound liver image sequence was developed to compensate for liver motion. Methods External surrogate respiratory motion signal is used to predict internal liver motion. An electromagnetic position sensor fixed on abdominal skin is introduced to track the respiratory motion, and 2D ultrasound images are used to track the liver motion synchronously. Based on these measurements, a subject-specific model describing the relationship of respiratory skin motion and internal liver motion is built and applied in real time (ultrasound-MRI/CT fusion imaging system) to predict and to compensate for the liver motion due to respiratory movement. Feasibility experiments and clinical trials were carried out on a phantom and eight volunteers. Results Qualitative and quantitative analyses and visual inspections performed by experienced clinicians show that the proposed model could effectively compensate for the liver motion, and the ratio of motion-compensated fusion error to the original varied from 10 % (0.96/9.40 mm) to 28 % (2.90/10.22 mm). Conclusions An online liver motion modeling and compensation method was developed that provides surgeons with stable and accurate multimodality fusion images in real time.