Image-guided therapeutic procedures such as focused ultrasound provide a game-changing solution for non-invasive cancer treatment. However, during non-invasive or minimally invasive procedures, surgeons have no direct interaction with the targeted area, and the success of the treatment relies entirely on the resolution of the images available from the targeted area. Currently, ultrasound imaging is used for this purpose; however, the required skills and precision remain challenging. To address this, I initiated a project at the Waterloo AI Institute, where I proposed the idea of AI-driven ultrasound imaging. This approach leverages AI for real-time labeling and identification of the targeted area and the volume treated. My role in this project involved leading a team of AI and medical imaging researchers to design and test the feasibility of this method. In addition to my leadership responsibilities, I was fully in charge of designing the hardware setup for the feasibility study and assisted the team in developing and verifying AI models. Our team successfully completed the project objectives, resulting in a new technology that enables physicians to monitor cancer treatment with quantitative data at the highest level of accuracy. . Our work was published in high-impact journals and received media coverage at the national and international levels. This project led to collaborations with Grand River Hospital and Focal Healthcare Inc., as well as international partnerships with the Korean Society of Thyroid Radiology. Additionally, our research efforts secured a $150,000 NSERC RTI Grant, enabling further advancements in this field.