ARRS Global Partner

British Institute of Radiology

The British Institute of Radiology (BIR), is an international membership organisation for everyone working in imaging, radiation oncology and the underlying sciences. We aim to support the work of our members and their colleagues to achieve professional excellence, provide continuing professional development for our multidisciplinary community, publish cutting edge research for our authors and readers across the world and influence and connect with the wider professional sector.

AI Pitfalls and What Not To Do: Mitigating Bias in AI

Various forms of artificial intelligence (AI) applications are being deployed and used in many healthcare systems. As the use of these applications increases, we are learning the failures of these models and how they can perpetuate bias. With these new lessons, we need to prioritize bias evaluation and mitigation for radiology applications; all the while not ignoring the impact of changes in the larger enterprise AI deployment which may have downstream impact on performance of AI models. In this paper, we provide an updated review of known pitfalls causing AI bias and discuss strategies for mitigating these biases within the context of AI deployment in the larger healthcare enterprise. We describe these pitfalls by framing them in the larger AI lifecycle from problem definition, data set selection and curation, model training and deployment emphasizing that bias exists across a spectrum and is a sequela of a combination of both human and machine factors.

Speakers

Judy Wawira Gichoya, MD; Kaesha Thomas, MD; Leo Anthony Celi, MD; Nabile Safdar, MD, MPH; Imon Banerjee, PhD; John D Banja, PhD; Laleh Seyyed-Kalantari, PhD; Hari Trivedi, MD; Saptarshi Purkayastha, PhD

How to Access

  1. Click on this link.
  2. Register and set up a password. (If you are already registered, log in with your email address and password).
  3. Follow the instructions to access the article.
  4. This article will be available until September 2024.

 

Featured Journal Publication

BJR is an international, multi-disciplinary journal covering the clinical and technical aspects of medical imaging, radiotherapy, oncology, medical physics, radiobiology and the underpinning sciences. BJR is essential reading for radiologists, medical physicists, radiotherapists, radiographers and radiobiologists.

BJR|case reports is the open access case report journal of the British Institute of Radiology and covers radiology, radiation oncology and all related sciences. To submit your work, please log onto our submissions system.

 

Global Exchange

BIR participated in an Annual Meeting Global Exchange featuring the United Kingdom during the ARRS 2024 Annual Meeting. During the Global Exchange, BIR representatives presented the featured course, Screening Patient Pathways Across the Pond: Highlights and Challenges for Radiology in the UK and USA with ARRS faculty. BIR President, Professor Stephen Keevil, received ARRS Honorary Membership.