4 R A D I O L O G Y From threat to essential ally Automation, not replacement: the true promise of AI in radiology Will artificial intelligence (AI) render radiologists obsolete? What seemed a likely scenario only nine years ago, has now given way to a quite differ- ent reality: At RSNA 2025, two experts outlined how AI tools are becoming essential allies – not replacements – for radiolo- gists facing an unprecedented workforce crisis. In 2016, Geoffrey Hinton, PhD, Nobel Laureate and widely known as “the godfather of AI”, told radiol- ogists attending the RSNA Annual Meeting that their jobs would soon be obsolete, replaced by artificial intelligence. Nine years later, radiol- ogists still dominate reading rooms, although AI tools have entered them. The dynamic, however, has changed. AI tools are now desper- ately needed to aid, rather than to replace, radiologists. How did this happen? Nancy Pham, MD, Assistant Professor of Radiology in neuroimaging and neurointerven- tion at Stanford Medicine, offered: “A large and unsustainable imaging vol- ume is outpacing the global baseline radiologist supply.” Speaking in a sci- entific session presentation at RSNA 2025 on AI’s expanding role in image interpretation, she outlined how increasingly heavy workloads and an ageing population of radiologists are driving rising turnover – radiologists leaving their positions or departing the profession entirely – with rates in the United States climbing from 5.3% in 2013 to 8.5% in 2022.1 In Europe, the situation is similarly bleak, with approximately 19% of radiologists retiring within five years.2 Three types of AI in clinical use Dr Pham explained that AI, used intelligently, can help minimize an impending crisis of too many diag- nostic imaging exams and not enough radiologists to read them. She identi- fied three viable applications: 1. Autonomous AI makes clinical decisions without human oversight – making it closest to the type of AI Prof. Hinton was referencing in 2016. Currently, two such products are in clinical use: LumineticsCore, the sole FDA-cleared autono- mous AI product, detects moder- ate-to-severe diabetic retinopathy, and either refers a patient to an eye care professional or advises that a repeat exam should be per- formed in 12 months. CE-marked Oxipit ChestLink reduces workload by identifying normal chest radio- graphs with 99.9% precision, allow- ing radiologists to focus exclusively on complex or abnormal cases. 2. Generative AI interprets free-text notes, laboratory results, vital signs, and a variety of reports. Dr Pham advised that she uses a generative AI tool at Stanford Medicine when a patient has a very complex and lengthy history. This tool rapidly generates well organized, detailed lists and prepares preliminary notes. 3. Augmented, or Assistive, AI serves as a co-pilot and enhances human decision-making. These AI tools handle non-interpretive and tedious tasks such as segmentation, volume measurement, and lesion tracking. Another application is to serve as a second reader acting as a safety net to reduce false positives and negatives. Breast imaging currently offers the strongest evidence of the benefits of AI as a second reader. ‘Noncontrast head CT scans are one of the highest volume exams ordered by emergency department physicians. Often performed daily for minor trauma, headache, or dizziness, the vast majority of these CT exams are normal,’ Dr Pham observed. An AI tool screening nor- mal studies would free radiolo- gists to focus on challenging cases and those with abnormal findings. ‘However,’ she cautioned, ‘the safety bars for such a tool are incredibly high. The brain is unforgiving.’ Tomorrow's highly automated reading room Tara Retson, MD, PhD, Deputy Chief of AI in Breast Imaging at UC San Diego Health, predicted that read- ing rooms will soon become highly automated. Multiple image-specific AI tools will integrate with PACS and electronic medical records, extract- ing information in seconds and gen- erating summaries of prior studies. Radiologists will have a wealth of information at their fingertips in sec- onds, she said. 'Real time alerts, triage, and prioritisation of unread exams is happening now. AI tools are advising which physicians need to be con- tacted immediately, what follow-up is needed, and whether it occurs. AI will standardize BI-RADS reports, do real-time coding, and embed guide- lines into reports.’ AI tools may also Looking back on two decades of education, collaboration and innovation perform control functions, such as performing drift monitoring tasks and audits for transparency. ‘AI fails in ways that we don’t expect’ ‘I believe that the biggest impact AI will have will be automating tasks, not performing activities that will replace radiologists,’ Dr Retson emphasized. ‘AI tools will promote patient centered care, empower radi- ologists to focus on complex imag- ing and cases by liberating them from tedious, time-consuming tasks, and improve workflow efficiency.’ However, vigilance remains crucial: ‘Human oversight is essential. AI must balance key concepts to augment clinical care. AI tools must be sub- jected to rigorous quality control. AI fails in ways that we don’t expect. No system is perfect. AI hallucinations are real,’ she warned. Special Report: Cynthia E. Keen m o c . e b o d a . k c o t s – f f o k n e d o r o G © References: 1. Parikh JR, Drake AR, Rula EY, et al. “Radiologist Turnover in the United States.” J Am Coll Radiol. 2026 Feb 25:S1546-1440(26)00031-1.; https://doi. org/10.1016/j.jacr.2026.01.009 2. Brady AP, Paulo G, Brkljacic B, et. al. “Current status of radiologist staffing, education and training in the 27 EU Member States.” Insights Imaging. 2025 Mar 15;16:59.; https://doi.org/10.1186/s13244-025- 01925-7 Nancy Pham, MD, is an Assistant Professor in neuroimaging and neurointervention at Stanford Medicine in Stanford, CA. Radiology of Tara Retson, MD, PhD, is an Assistant Professor of Radiology at UC San Diego Health. Her research focuses on deep learning applications in medical imaging. ESOR at 20: building a common language for European radiology Past and present leaders of the European School of Radiology (ESOR) reflected on the evolu- tion of radiology education in Europe and the challenges fac- ing the specialty at ECR 2026 to mark the 20th anniversary of the school. Established in 2006, ESOR was created at a time when radiology education across Europe remained highly heterogeneous, with major differences in access to subspecial- ty training, research opportunities, and international mobility. Twenty years later, the school has grown into a broad educational network spanning Europe and beyond. Harmonising education across and beyond Europe For Prof. Nicholas Gourtsoyiannis, founder and former Scientific and Educational Director of ESOR, the school was initially conceived as a response to the need for a more harmonised educational structure within European radiology. ‘ESOR was the fulfilment of a com- mon European dream towards radiological education,’ he recalled during the anniversary discussion at ECR. This international dimension was also highlighted by Prof. Valérie Vilgrain, former Scientific and Educational Director of ESOR, who described the diversity of back- grounds and medical cultures with- in the school as one of its greatest strengths. ‘We work with radiologists not only from Europe, but all over the world,’ she said. ‘This is very different from one country to another, and it is so rich to work with people with a different environment.’ Going digital after Covid Vilgrain also reflected on the impact of the Covid-19 pandemic, which accelerated the development of online learning and hybrid formats, changes that have since become a permanent part of the school’s activities. ‘Now ESOR is probably stronger because we have faced that,’ she noted, explaining how the crisis pushed the organisation to reassess priorities and educational struc- tures. Among the initiatives developed during that period were the online foundation courses covering major radiology subspecialties, which significantly broadened access to high-level radiology education. Prof. Christian Löwe, current Scientific and Educational Director of ESOR, described the combina- tion of online education and onsite interaction as one of the key evo- lutions of the school over recent years. ‘Many things can be done better face to face,’ he said, while acknowl- edging the role digital learning now plays in improving accessibility and international participation. Growing collaborations Partnerships also emerged as a central theme throughout the inter- views. ESOR today collaborates with national radiological societies, academic institutions, subspecialty organisations, and industry part- ners to support fellowships, schol- arships, and educational initiatives. For Löwe, maintaining educational standards increasingly depends on this collaborative approach. ‘We really believe that we can achieve harmonised excellence in education only through partner- ship,’ he explained during a sepa- rate interview focused on coopera- tion between academia and indus- try. The discussions further reflected the broader technological transfor- mation affecting radiology itself. Artificial intelligence, advanced imaging technologies, and data-driv- en medicine are reshaping clinical practice while creating new educa- tional demands for radiologists. of Fulvio Renoldi Bracco, CEO of Bracco Imaging, highlighted the importance collaboration between industry and the medical community as innovation acceler- ates. ‘Without a strong partnership with the physician community, that would not be possible,’ he said. Looking ahead, ESOR’s leadership outlined several priorities for the coming years, including expanding fellowship opportunities, strength- ening cooperation with radiolog- ically under-resourced countries, and developing new educational formats adapted to radiologists at different stages of their careers. Twenty years after its creation, ESOR reflects the broader evolu- tion of radiology itself – a spe- cialty increasingly defined by con- tinuous education, international collaboration, and technological adaptation. By Mélisande Rouger EUROPEAN HOSPITAL Vol 35 Issue 2/26