Grant: Discovery Grants Program - Individual

Background: Primary liver cancer is the third leading cause of cancer death and the sixth most diagnosed cancer worldwide. Hepatocellular carcinoma is the most common subtype of primary liver cancer, but the liver is also a frequent site for metastatic cancer originating from other body parts. Surgical resection and liver transplantation are being used as curative options for patients with early stage liver cancer. Unfortunately, only 10-20% are suitable candidates for these approaches. Furthermore, postoperative complications are observed in up to 33% of patients undergoing surgical resection and liver transplantation, respectively. Therefore, other treatment alternatives, such as percutaneous liver ablations are used to overcome or mitigate the limitations of surgical resections and liver transplantation. Need for image-guided intervention: Percutaneous liver ablations, such as radiofrequency and microwave ablation, offer curative treatment options for early-stage liver tumours, particularly for patients who are not eligible for surgery. CT guidance is the standard-of-care imaging modality used in the ~2hr-long procedure, but CT's wide availability and accessibility cannot always be guaranteed in many centers, especially in cost-constrained healthcare systems. Ultrasound image guidance: 2D ultrasound (US) provides real-time imaging, is widely available, and is inexpensive. But, 2D US image guidance is subjective, relies on a high level of experience, and one or multiple 2D US images cannot provide sufficient measurement information, such as the tumour size, shape, and the ablation needle position. Our technological solution: We have invented 3D US technologies and extended their use by developing a prototype mechatronic 3D US-based liver ablation guidance system. It reduces operator dependence, allows obtaining a larger field-of-view of the liver, and provides more accurate quantitative measurements, such as tumour size and shape. However, our system lacks several technological components critically needed to make the system useful. Therefore, our overall research goal is to continue to develop critically needed system components and test our 3D US-guided system with phantoms to ensure that detection of tumours and needle guidance are accurate by: (a)developing an image registration of diagnostic CT images with the intra-procedural 3D US images to improve tumour visibility for cases with ultrasonically invisible or poorly visible tumours for better tumour identification and localization, (b)developing a machine learning algorithm used to localize, segment, and track the therapy needles in real-time to improve real-time needle identification for accurate US-based needle guidance, and (c) testing the complete prototype 3D US-guided system in the laboratory using liver phantoms, which would be imaged by CT and 3D/2D US, for evaluation of the identification of the tumours, needle guidance trajectory information, and needle tip locations.