Computer Vision for Medical Imaging Applications
"The tutorial was featured on the BEST OF ECCV of Computer Vision News "
Abstract
Medical imaging plays a crucial role in modern medical diagnosis and intervention, cross-cutting all clinical specialties. However, the continual rise in medical imaging has not been matched by the number of radiologists trained to interpret the images [1]. Meantime, AI has shown exciting potential to assist human readers in vision tasks to achieve better performance [2], creating an opportunity for computer vision algorithms to support medical experts to deliver affordable healthcare at scale.
In this tutorial we will give an overview of the medical image analysis domain - the different types of medical scanner, how they work, and what they are used for - and the challenges and solutions required when translating computer vision techniques from academia to real-world deployment.
[2] Wu, N., Phang, J., Park, J., Shen, Y., Huang, Z., Zorin, M., Jastrzębski, S., Févry, T., Katsnelson, J., Kim, E. and Wolfson, S., 2019. Deep Neural Networks Improve Radiologists' Performance in Breast Cancer Screening. IEEE Transactions on Medical Imaging. doi: 10.1109/TMI.2019.2945514
Tutorial Topics
Schedule of talks:
How AI will transform healthcare: A clinician’s view (Keith Muir)
Introduction to medical imaging (Keith A. Goatman)
Challenges in medical image analysis (Keith A. Goatman)
Case studies in deployment of medical AI (Alison Q. O'Neil)
Beyond imaging: Multimodal AI using the whole patient record (Alison Q. O'Neil)
Future opportunities in medical image analysis (Alison Q. O'Neil)
Videos available to ECCV attendees at: https://workshopsandtutorials.eccv2020.eu/papers/subject/computer-vision-for-medical-imaging-applications/
Organisers
Alison O’Neil (EngD)
Canon Medical Research Europe & University of Edinburgh (Hon. Fellow)
Alison O'Neil is a Senior Scientist in the AI Research Team at Canon Medical Research Europe and Honorary Research Fellow at the University of Edinburgh. She leads a team of scientists and research students working on machine learning techniques for healthcare applications for medical imaging, natural language processing, and electronic health record data. Her research has covered techniques for medical image registration, segmentation of anatomy and pathology, anatomical landmark detection, and more recently prediction of outcomes from clinical data and the extraction of semantic information from medical text. She is Associate Editor of the IEEE Journal of Biomedical and Health Informatics and is the main organiser of the ICLR 2020 “AI for Affordable Healthcare” workshop. She has multiple publications and patents in the domain of medical AI.
Keith Goatman (PhD)
Canon Medical Research Europe & University of Aberdeen (Hon. Fellow)
Keith Goatman is a Principal Scientist in the AI Research team at Canon Medical Research Europe, and Honorary Research Fellow at the University of Aberdeen. He has experience in clinical, academic, and industry-based image analysis and machine learning research. He currently leads a team of scientists and research students working on machine learning techniques for healthcare applications involving both imaging and non-imaging health data. His computer vision adventure began in 1994, writing software to measure valves within veins using fibre optic imaging. His doctoral work at the University of Sheffield explored automated ultrasound image characterisation of carotid artery plaques to predict stroke risk. He has wide experience of medical imaging modalities including nuclear medicine and PET, ophthalmic imaging, microscopy, CT, and MRI. He taught for 10 years on the University of Aberdeen Medical Physics Masters’ programme. He has 30 journal publications and has several patents.