Speakers - CIRWC 2024

Abstract

Contrary to popular belief, Machine Learning (ML) as we know it is not something “modern”, in fact the term “machine learning” was first used in 1959. So how come machine learning has suddenly come into lime-light? The first reason is High Performance Computing (HPC) and Massively Parallel Processing (MPP), both have been around since 2000’s and 1990’s respectively. The second reason is availability of this immense processing power on cloud, such as Amazon Web Services (AWS). The applications of this powerful triad in cancer interventional radiology treatment are numerous. However, in this paper we will restrict ourselves to seven major applications i.e. Image Analysis and Segmentation, Computer-Aided Diagnosis (CAD), Treatment Planning and Simulation, Procedural Guidance and Navigation, Predictive Analytics and Risk Assessment, Quality Assurance and Workflow Optimization and Clinical Decision Support.

The extent to which a healthcare practitioner should know about machine learning (ML) depends on various factors, including their specific role, area of expertise, and interest in incorporating ML into their practice. Here are some of the key considerations we will discuss in this paper i.e. Awareness, Data Literacy, Clinical Relevance, Collaboration and Ethical and Regulatory Considerations.

Some regard ML as a “tool-set”, however, ML is much much more as compared to traditional tool-set due to Variety of Tools, Specialized Functions, Flexibility and Adaptability, Skill and Expertise, Continuous Improvement and Problem-Solving Capability; all of this will be addressed in this paper. In this paper we will consider each of the major applications of ML in cancer interventional radiology treatment and identify and explain which techniques and “tools” to be used for particular application with justification. For example, for Procedural Guidance and Navigation, the techniques of choice are i) Image Registration and Fusion ii) Image Segmentation and Visualization iii) Needle and Catheter Tracking Algorithms iv) Augmented Reality (AR) and Virtual Reality (VR) v) Robotics and Automation vi) Predictive Modelling and Feedback Control and vii) Intraoperative Decision Support Systems. The paper finally ends with conclusions and list of references.