Carlo Catalano took up the cudgels for using artificial intelligence (AI) in radiology at the high-profile event “European Cancer Imaging Initiative – Joining forces for AI-powered imaging to save lives” on February 6 in Brussels, Belgium. He emphasized that AI is an indispensable support for every radiologist, but also made it clear that targeted investments, better European networking, and a comprehensive range of training courses are required to exploit the full potential of this technology.
Catalano pointed out, that the increasing amount of image data and the growing need for precise diagnostics have led radiologists to work increasingly in teams – together with radiology assistants, physicists, administrators, and also with artificial intelligence (AI). “AI can evaluate image data faster and enable the transition from purely qualitative to quantitative image analysis. This allows diseases to be detected earlier, therapy successes to be assessed more precisely, and the basis for personalized medicine to be created,” said the Chairperson of the European Society of Radiology (ESR) Board of Directors.
Integrated diagnostics
Linking imaging, laboratory diagnostics and genomic data is just as important. However, efficient use of this data can only be achieved through powerful AI platforms that analyze enormous amounts of data and process it for clinical decisions, Catalano summarized. This is precisely where the EUCAIM (European Cancer Imaging Initiative) project comes in. It networks imaging data from various European countries, making it usable for research and clinical applications. “The ESR strongly supports this approach,” emphasized Catalano.
Despite this progress, there is still a backlog in the IT infrastructure of hospitals in many European countries. Without targeted investment in data security and networking, the potential of AI and big data would remain untapped: “Hospitals must be put in the condition that they can provide very secure data to these federated infrastructures. Otherwise, it becomes difficult for many of our members, for ourselves as radiologists, and for healthcare practitioners.
Lack of knowledge about AI
Another problem is the lack of knowledge many doctors have about AI technologies. While radiologists often already have a high level of AI expertise compared to other specialties, many other doctors lack knowledge about the benefits and limitations of these technologies. Catalano is therefore calling for targeted training and support programs: “It’s absolutely necessary that there are enough funds to spread the information, to train and to educate physicians.”
Reducing inequalities in care
According to Catalano, AI could not only improve diagnostics but also reduce inequalities in medical care within Europe. In many countries and even within individual regions, there are considerable differences in the availability and quality of radiological examinations. AI could help to close these gaps. However, the success of these technologies requires interdisciplinary cooperation between doctors, researchers, developers, administrators, and political decision-makers. The expert appealed to the European Commission to create more incentives to bring scientific societies, hospital operators, and the IT industry closer together.
The event was organized by the European Commission, the European Society of Radiology, and EUCAIM.
