Join us in Chicago for RSNA 2021. Whether you are attending in-person or virtually, we welcome you to stop by our booth, attend our presentations, and book a meeting to learn more about our award-winning solutions for breast cancer screening.
Monday 29 November, 12:30pm (CST)
AI Theatre: Safety nets in Medical Imaging AI – What they are and why they matter
Dr. Ben Glocker – recently with Imperial College London and now Kheiron’s Head of Machine Learning Research
Safety nets provide a margin of protection or security against AI failures. In medical imaging AI, we have a responsibility to make sure that the use of AI is safe and that any predictions made by an algorithm are trustworthy.
In this session we will discuss various safeguards, including automatic quality control, failure detection, and uncertainty estimation. We will also delve into robustness and reliability in the context of a mismatch of data between the method development and clinical deployment, and we will explore the use of causal reasoning and how it can help identify potential biases.
Tuesday 30 November, 12:15 (CST)
Lunch & Learn: The importance of inclusivity and diversity in healthcare AI
Dr. Bonnie Joe, MD, PhD – Chief of Breast Imaging, UCSF
Dr. Amy Patel, MD – Medical Director of The Breast Care Center at Liberty Hospital
Dr. Hari Trivedi, MD – Assistant Professor of Radiology and Biomedical Informatics, Emory University
Facilitated by Sarah Kerruish – Chief Strategy Officer, Kheiron Medical Technologies
AI applications are developed on vast amounts of data. In healthcare, the quality of that data, and the types of data being used to train the algorithm, are critical. Whilst AI isn’t intrinsically biased, if the data used to create these algorithms doesn’t contain a diverse representation of society in terms of ethnicity, age, geography and gender, they cannot produce accurate, generalisable outputs.
For example, the incidence rate of breast cancer for Black women is close to that of white women. However, the mortality rates are markedly different, with Black women having a 40% higher death rate from breast cancer, according to data compiled by the American Cancer Society. Possible risk factors for this include socioeconomic status, biological and genetic differences in tumours and differential access to healthcare. AI has the potential to revolutionise healthcare but precautions must be taken to ensure we are eliminating rather than perpetuating these kinds of biases.
Join this session to hear from leading experts as they discuss the importance of generalisability in healthcare and how the rapid adoption of AI means that it is critical that stakeholders take steps to address algorithmic biases now.
Due to COVID regulations, RSNA requires that all Lunch & Learn attendees pre-register. Click on the button below to reserve your seat now!
Throughout the week
Join us at Booth #4738 for a variety of presentations throughout the week. Topics include:
- Our vision for the future of cancer detection (3:00 Monday)
- How AI can help address the COVID-related breast screening backlog (11:00 Tuesday)
- Results of our ground-breaking mammography AI clinical study (11:00 Wednesday)
The full schedule will be published soon.