This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.
In conversation with Professor Erik Ranschaert, MD, PhD: Next generation radiology at ECR 2024
The European Congress of Radiology 2024 is about to begin. As we embark on this experience of exploration and discovery, we introduce an insightful conversation with Dr Erik Ranschaert, Visiting Professor of Radiology at Ghent university, and advisor of Quibim. Renowned for his expertise in radiology, Dr Ranschaert brings his invaluable knowledge to our conversations.
His unique insights and comprehensive analysis will significantly enhance our understanding of the current trends, advancements, and challenges within the Medical imaging landscape.
Latest advancements
Could you summarize the current state of AI applications in X-Ray examinations and how they transform traditional radiology? What advancements will be highlighted in this field at ECR 2024?
In the past 5-10 years, a significant evolution has occurred in the availability of many solutions for different modalities, mainly in Neuroradiology, Chest Radiology, Mammography, and musculoskeletal radiology. There is a substantial growth of clinical applications in radiology departments and hospitals. In scientific literature, we can notice a gradual shift towards studies evaluating AI in a clinical “real world” environment, which will contribute to the growth in trustworthiness. I guess that we will also see a growing number of paper focusing on providing advice and “guidelines” regarding the management of AI in a clinical environment. A good example is the recently published multisocietal paper with practical considerations on developing, purchasing, and monitoring AI tools in radiology. Furthermore, a “boost” was given recently to the AI-relationship by the release of Large Language Models (LLM) and vision-language or Multimodal Foundation models. In other words, the relationship between AI and radiology is growing and gradually becomes pre-marital, but we should consider rethinking the “definition” of such marriage. I will explain this in more detail during my lecture at ECR: ‘AI and X-ray: the perfect wedding.’
Noteworthy benefits
Considering your extensive experience integrating AI in medical imaging, what are AI’s most significant benefits to radiologists and the overall diagnostic workflow? How does it enhance efficiency and diagnostic accuracy?
Each clinical setting has its own unique needs and AI solutions should be chosen based on the most relevant bottlenecks and problems. There is a shortage of radiologists in numerous countries to interpret all radiological examinations, making AI a valuable asset in enhancing diagnostic services under these circumstances.
AI has been shown to significantly improve the quality of examinations such as breast and lung cancer screening. Assistance by AI can significantly reduce the workload of the radiologists involved, as well as reduce the costs associated with organizing such large-scale screening initiatives.
Obstacles to consider
From your perspective as a leader in radiology and an advisor to AI start-ups, what do you identify as the most critical barriers to successfully implementing AI technologies in radiology departments?
Many radiologists still need to be made aware of the potential advantages. The existing distrust can be reduced by having radiologists work with an AI solution in their practice. Another major barrier is funding and the difficulty of making a good business case, with a clear ROI. A balance must also be sought between quality improvement and production increase or workflow improvement, and this can be challenging. In Europe, where most healthcare systems are based on a form of socialized medicine, I believe it is also the government’s task to make a constructive contribution here so that the applications can be enabled on a larger scale.
The goal of redefining radiology
As someone who has co-authored the first book about AI for Medical Imaging, you have a deep insight into the evolving landscape of radiology. In the context of ‘Next Generation Radiology,’ are there any emerging concepts or technologies you believe are set to redefine the future of diagnostic radiology?
The development of LLMs and multimodal foundation models will open the door to the further development of personalized medicine (precision medicine), especially since many more data sources can be utilized and combined for developing new AI-solutions. Consecutively it will be possible to, develop and implement AI applications that are much better tailored to the users and their treatment.
It is necessary however to create tools and platforms allowing to monitor the performance of algorithms on the studies performed and to systematically evaluate the outcomes of therapy in patients, especially in the domain of oncology.
The alliance of radiologists and technology
How do you foresee the evolution of partnerships between radiologists and technology developers to further diagnostic precision and operational efficiency?
I see an increase in cooperation between these parties. They are learning to communicate with each other in a common language, which is only possible in an environment where cooperation is stimulated. Academic or research centers should focus on such collaboration, and progressively more initiatives are taken to move in that direction. I am also involved in several projects in which AI research is done in a multidisciplinary group, which is very interesting and learnful.
Referents that set the path for next generation radiology
As a professor and an influential figure in radiology, how do you incorporate your insights on AI and machine learning into your teaching and mentorship, aiming to prepare the next generation of radiologists for the digital revolution in healthcare?
Over the past few years at Ghent University, I’ve brought up topics on AI for medical students about which they can then do a bachelor’s or master’s thesis. This is a very rewarding process. Seeing how these students proceed is very promising. It also confirms that there is a definite interest in these topics, and that the new generation of physicians in training is undoubtedly willing to learn how to deal with them and make meaningful use of them. At the EuSoMII society, of which I have been president,We have succeeded in bringing together a large “young club” consisting of radiology residents, young radiologists, and students from other disciplines, with one common interest: AI development and usage in radiology and medicine. So, it is being worked on!
Radiologists have a critical role to play in transforming medicine as we know it. At ECR, innovation and discovery converge to set new standards for patient care. This motto resonates with our commitment to unlocking the full potential of radiology to improve patients’ lives.
This enriching conversation with Dr Erik Ranschaert underscores the transformative influence of AI in precision medicine, highlighting its capacity to transform healthcare. Join us at ECR to explore the ongoing advancements continually redefining medical imaging.