2023 AI Showcase Highlights: Innovations & Insights

Author: 

Segmed Team

Reading time / 
5 min
Company News

Generative AI for Synthetic Medical Imaging
What can we expect?

Artificial intelligent (AI) models for medical imaging tasks, such as classification or segmentation, require large and diverse datasets of images. However, due to privacy and ethical issues, as well as data-sharing infrastructure barriers, these datasets are scarce and difficult to assemble.Synthetic medical imaging data generated by AI from existing data could address this challenge by augmenting and de-identifying real imaging data. In addition, synthetic data enable new applications, including modality translation, where a CT scan can be transformed into a corresponding MRI; contrast synthesis, where a contrast agent is synthetically added to the image; and professional training, where synthetic images of rare cases can improve a trainee’s clinical knowledge. However, synthetic data also poses technical challenges, such as ensuring the realism and diversity of the synthesized images, evaluating the performance and generalizability of the models trained on synthetic data, and high computational costs.In this presentation, we provide an overview of the clinical applications, technical principles, and prospects of synthetic medical imaging data.

Speaker Information

Martin Willemink

CEO & CofounderSegmed

Martin is a physician, epidemiologist, biomedical engineer, scientist, and entrepreneur. He is the CEO and co-founder at Segmed - the fastest growing self-serve platform for diverse, high quality, de-identified medical imaging data. Previously, Martin led the clinical cardiovascular imaging research at Stanford Radiology.Martin has published 100+ peer-reviewed papers, has been invited to present 30+ lectures at international scientific meetings, is a Fulbright laureate, and received funding from the American Heart Association, Philips Healthcare, Stanford University, and more. He is designated as a Fellow of the Society of Cardiovascular Computed Tomography (SCCT) and is an active member of multiple European and North American clinical scientific societies. His work has amongst others been recognized with awards from the Radiological Society of North America, European Society of Radiology, Society of Thoracic Radiology, and European Society of Thoracic Imaging.

Brad Genereaux

Global Lead, Healthcare AlliancesNVIDIA

As a passionate and engaged healthcare and medical imaging leader, Brad's mission in life is to evangelize the ubiquitous adoption and integration of seamless healthcare and medical imaging workflows into everyday clinical practice, driving resilient turnkey solutions at scale. With more than 20 years of industry experience in hospital, radiology, and enterprise imaging IT, his focus is to accelerate artificial intelligence and deep learning, visualization, analytics, and virtualization solutions within the medical imaging and smart hospitals domain. As the Global Lead of Healthcare Alliances at NVIDIA, Brad is at the forefront of imaging interoperability within the healthcare ecosystem and is deeply involved with the community in building and implementing standards like DICOM, HL7, and IHE.