MICCAI 2019

MICCAI 2019 DAILY 8 EXPO technologies is crucial. Everything in AI is cutting-edge and it’s important to stay at the frontier of that. She started working at GE two years ago, when they were working on their first AI-based feature recognising standard image views from ultrasound exams. Although it was only a couple of years ago, this feature is already on the market and in the hands of doctors in hospitals around the world improving their workflows. Kristin found that turnaround time particularly motivating and says they are highly driven to work fast and provide for customer needs as soon as they possibly can. In terms of what’s coming up next, one step is to automate analysis of images. They also want to work on the acquisition of images themselves – which is a tough challenge in ultrasound – and going forward into automating measurements because these measurements drive diagnosis. Once you automate all The AI / Data Science team of Cardiovascular Ultrasound Kristin McLeod is the AI and data science team leader for cardiovascular ultrasound at GE Healthcare. She is joined at MICCAI by colleagues from Women’s Health Ultrasound, Acute Care, Magnetic Resonance, Cardiovascular Ultrasound and their Global Research Center. GE Healthcare are a gold sponsor and ran a workshop yesterday called Smart UltraSound Imaging. Kristin speaks to us about their work. Throughout GE Healthcare, they are building up their AI offering, as they see the potential for AI to make a significant improvement to clinical workflows. It will make measurements more robust and make it easier for users to use their systems. It will also make things more comfortable for patients, by making examinations faster, and enabling physicians to do more measurements, thus providing more information for them to make diagnoses. Ultrasound is known to be a bit subjective and user-dependent, so they want to use AI to remove this subjectivity and enable more robust measurements by standardising the acquisition and analysis of ultrasound images. They have a lot of data readily available. Kristin’s team counts around 10 people, including many students. She was a researcher herself before, so being familiar with the latest