AI in Medical Imaging Research Seminar – Cathal McCague – "Stop rolling the DSC! - A new framework for assessing the clinical utility of segmentation algorithms"

Add to your list(s) Download to your calendar using vCal

• Cathal McCague, Department of Radiology
• Tuesday 27 September 2022, 10:00-11:00
• Berridge room on L5 at the Department of Radiology, Addenbrooke's hospital.

If you have a question about this talk, please contact Ines Machado.

The Cancer Research UK Cambridge Centre and the Department of Radiology at Addenbrooke’s are pleased to announce the launch of a seminar series to support campus-wide efforts to engage with the fast-developing field of Artificial Intelligence (AI) applied to medical imaging. The seminars will take place on a regular basis (monthly) with the aim to exchange information and ideas on AI applied to imaging.

The first event will take place on the 27th of September at 10am in the Berridge room on L5 at the Department of Radiology, Addenbrooke’s hospital.

Directions to the venue can be found here.

This is a hybrid event so you can also join via Zoom

Meeting ID: 990 5046 7573 Passcode: 617729

We are delighted to introduce our first speaker Cathal McCague. Cathal is an honorary radiology registrar based at Addenbrooke’s Hospital and PhD student working under the supervision of Dr. Mireia Crispin-Ortuzar. His research focuses on the development and validation of AI-based segmentation tools for ovarian cancer on CT scans. Additionally, he is interested in the wider implications of the application of such AI-based technologies into the clinical setting. He collaborates with Dr. Charlotte Tulinius from the Department of Public Health and Primary Care on research to understand the perspective of ovarian cancer patients on the use of AI tools for radiological image interpretation. Cathal will present work related to the development of a new framework for appraising the performance of segmentation models, which assesses practical value in a clinical setting and provides a qualitative evaluation of utility.

We hope to host seminars on a monthly basis during term times open to all Biomedical Campus staff, students and affiliates. The seminars are intended for a multidisciplinary audience. Presentations last approximately 30 minutes, after which there is plenty of time for questions and discussions. If you have any questions or are interested in presenting your research work, please email Dimitri Kessler (dak50@cam.ac.uk) or Ines Machado (im549@cam.ac.uk).

This talk is part of the Cambridge MedAI Seminar Series series.

This talk is included in these lists:

• All Talks (aka the CURE list)
• Berridge room on L5 at the Department of Radiology, Addenbrooke's hospital
• CCC talks for website
• Cambridge Centre for Data-Driven Discovery (C2D3)
• Cambridge MedAI Seminar Series
• Cambridge talks
• Chris Davis' list
• Interested Talks
• Interested Talks
• Trust & Technology Initiative - interesting events
• bld31
• ndk22's list
• ob366-ai4er
• rp587

Note that ex-directory lists are not shown.