Liver Biopsy Sampling System Guided by the Internal Organs Motion and Ultrasound Imaging
Date
2020-08-22
Authors
Laila Ahmad Omar Abbasi
ليلى أحمد عمر عباسي
Journal Title
Journal ISSN
Volume Title
Publisher
Al-Quds University
Abstract
Percutaneous needle access is a critical step of many clinical procedures, such as injections, diagnostic biopsies, and cancer ablation treatments, the higher the accuracy of needle placement and trajectory planning, the better the procedure results. However, the desired optimal needle targeting and positioning are challenging due to physiological respiratory motion and tissue deformation, which can cause mistargeting, resulting in needle placement error, ineffective delivery of treatment, or mistaken diagnosis. Advancement in medical technology and the development of image-guided navigation systems enable better interventional processes and higher procedure accuracy and efficacy. However, despite the wide variety of strategies and technologies implemented in needle placement and navigation systems, clinically applicable solutions are still lacking due to incompetent systems of addressing all technical and patient-specific challenges associated with image guidance interventions. The literature discussed the major effect of respiration on organs movement, and reported a high magnitude of organ displacement during breathing, specifying the organ motion as a major source of lesion targeting error in needle interventions. Although, the state-of-art image-guided navigation systems are not capable of compensating and dealing with this error factor.
The development of a precision image-guided navigation system that accounts for respiratory induced lesion displacement, integrated with medical imaging processing and computer vision techniques is proposed in this thesis. Aiming to control needle placement sources of error in clinical applications of biopsy sampling in a process that adapt to the current clinic workflow. The system provides physicians with preoperative feedback on needle biopsy sampling plan, allows for patient-specific biopsy plan, target tracking, performs image processing, volume rendering of targets region of interest, needle insertion, and extraction considering organ movement.
The system implemented MOSSE tracker for target region tracking during the intervention, and has been evaluated on 9 patients’ data with 27 simulated liver tumor biopsy sampling procedures, and showed a reliable tracking with a high system accuracy with a mean overall
error of 1.78 ± 0.8 mm. The results show that controlling needle insertion based on the motion improved the targeting accuracy and made possible for future critical clinical applications