Artificial Intelligence has emerged as the next step in the evolution of the medical imaging industry. At the Medical Imaging Convention we recognise these advancements and will put them on centre stage for all visitors to experience first hand. Find out everything about this new focus area here, and register for tickets by following the link at the top of the page.
Roche, known for its biopharmaceuticals and in-vitro diagnostics solutions has entered the field of Clinical Decision Support by building a portfolio of software products. Learn how the NAVIFY ® Decision Support portfolio leverages excellence in clinical workflow, advanced analytics and innovative strategic partnerships to bring together diagnostic, real-world data to deliver personalized healthcare.
Artificial intelligence has the potential to change the way we treat patients but the ethical implications are complex. This lecture will highlight some of the key areas that the clinical community needs to think about.
Artificial Intelligence is everywhere, or appears to be. Yet whilst it has the potential to make a big difference in healthcare, in truth there are still relatively few genuine AI deployments that genuinely impact patient outcomes. Based on two decades of experience commercialising AI in healthcare, Anthony will provide some ‘do’s and don’ts’ to consider when bringing AI algorithms
to market, including the most important: “Get out of the building!”.
In this talk, I will present the research projects that we have been working at Imperial College London and HeartFlow Inc. The presentation will mainly focus on applications of machine learning methodologies in medical image analysis tasks, including cardiac CT/MR image reconstruction, image quality enhancement, and image segmentation. In particular, I will present a few use cases of deep neural networks in clinical workflows, which are utilised to automate quantitative measurements and leverage the knowledge learnt from large annotated medical datasets.
An industry prospective of the challenges and pitfalls in developing and deploying deep learning algorithms within the NHS breast cancer screening programme.
AI hold tremendous promise for healthcare. Technologists frequently talk about ‘disrupting’ healthcare in the same way that technology has disrupted retail, communication, entertainment, and many other spheres of life. We are now seeing a culture clash, between the ‘move fast’ world of technology, and the ‘safety first’ world of medicine. What is the role of regulation in striking the right balance? What’s on the horizon? What should we, as patients and as clinicians, expect?
Large-scale, properly curated and live databases are essential for the creation and validation of artificial intelligent-enabled systems such as Computer Aided Detection/Diagnosis. It is not only important to ensure visibility of datasets used for training and testing, but also vital that validation of emerging AI-systems is possible utilising independent validation sets. This seminar will discuss the practicalities of forming such large-scale, drawing upon experience gained forming the OPTIMAM image database.
Medical diagnosis is a complex and time-consuming process. Facilitating it could bring us a number of benefits, e.g. increase the chances of early detection of cancer. During the seminar we will see how current state-of-the-art computer vision algorithms can be used to achieve this goal.
I will review the importance of breast density assessment for cancer risk prediction, identifying women for whom mammography is less effective, and assessing the efficacy of risk reducing interventions. I will then discuss current methods for density assessment, looking at the strengths and limitations of different approaches. Finally, I will describe density assessment using convolutional neural networks, including a review of the issues associated with developing and validating such approaches.
Clinical decisions in acute stroke are few: to thrombolyse or not? how about thrombectomy?; and, in grave cases, should we surgically decompress? However, the myriad factors – particularly imaging - which clinicians weigh up to make these decisions are anything but simple. AI methods may represent the optimal way by which multi-stream data can be quantified, rationalized and brought to bear on critical questions of management and outcome.
This seminar will look at the evolving field of Artificial Intelligence and the potential for use within the NHS. It will particularly focus on how the Clinical Safety can be assessed and the evolving policies and standards that should be used.
