Chair: Professor Timothy Cootes, Professorial Research Fellow, Division of Informatics, Imaging & Data Sciences, University of Manchester
An overview of the Radiology AI market: past, present and future
Dr Jamie Chow, Clinical Lead at Blackford, a UK health tech company that has developed a platform for the integration and delivery of AI applications within Radiology
The talk will discuss how the recent boom in medical imaging AI came about, the current state of the AI market and the direction in which it is going.
- Have a broad understanding of the evolution of the radiology AI market
- Appreciate the common types of AI applications available today.
- Understand the emerging trends within radiology AI.
AI in chest imaging: Experience from the Greater Manchester chest x-ray AI project
Dr Rhidian Bramley, Clinical Lead for Diagnostics, Digital and Innovation, Greater Manchester Cancer
Lung cancer is the leading cause of cancer death in the UK and worldwide. Chest x-ray (CXR) is the key initial investigation for patients with chest symptoms. This talk will look at the challenges of reporting CXR in cancer diagnosis and experience from the pan Manchester chest AI project to assess how AI can assist in CXR interpretation leading to faster diagnosis of patients with lung cancer.
- Understanding the limitations of CXR in cancer diagnosis.
- How to overcome challenges in implementing AI at scale
- Recognising where AI is most likely to help in this step in the cancer pathway.
AI in mammography
Professor Sue Astley, Chair Division of Informatics, Imaging & Data Science, University of Manchester
The talk will cover new developments in the use of AI in mammography, including applications assessing breast density and risk, and software for triage and early detection. Strengths and limitations of AI-based approaches will be discussed.
Radiomics and radiogenomics
Dr Anubhav Datta, post-doctors researcher at The Christie NHS FT with an interest in cancer imaging and developing quantitative biomarkers for use in clinical trials
Imaging data is incredibly rich and complex, and this is why imaging analysis models developed over the past 3 decades have delivered only a handful of validated quantitative biomarkers. Radiomics has shown early promise as a way to characterise in vivo tumour heterogeneity however has poor overall precision and unknown biological accuracy. A global initiative on imaging biomarker standardisation aims to address many of these concerns.
- To understand what radiomics research and why it exists.
- To improve the attendees' critical evaluation of radiomics research.
- To identify potential uses of radiomics/radiogenomics in tomorrow’s healthcare
The use of chatbots in education and research
Dr Katy Szczepura, Associate Professor in Medical Imaging Physics, University of Salford
There has been a surge of interest in Chatbots, or generative AI, for use in education and research. This session will discuss the benefits, risks and appropriate use of generative AI, and will cover ethics, equity, justification, intellectual property and critique.