Artificial Intelligence (AI) is reshaping the future of radiology - enhancing diagnostic accuracy, reducing variability, and easing clinician workload. In this exclusive interview, Prof. Dr. med. Mathias Goyen, Chief Medical Officer of GE HealthCare, shares how AI is transforming healthcare delivery by complementing radiologists, streamlining workflows, and driving global health equity.
AI in radiology is not a replacement for the radiologist, it’s a powerful partner. The goal is to augment human intelligence, not substitute it. AI algorithms can sift through vast datasets rapidly, flag abnormalities, and prioritize cases, allowing radiologists to focus their expertise on complex interpretations and patient-centric decisions. It’s a symbiosis: machines bring speed and consistency, while radiologists bring context, clinical reasoning, and empathy. Together, they form a more powerful diagnostic team.
AI is improving both the “how” and the “how fast” of diagnosis. For instance, research shows that algorithms can now detect early-stage cancers, such as breast, lung, or prostate, with remarkable precision, sometimes even before they’re visible to the human eye. AI can also automate time-consuming tasks like segmentation, measurements, and report generation. One of the most transformative developments is research in triage in emergency settings. Research is underway for AI to automatically flag critical findings such as intracranial hemorrhages or pulmonary embolisms, expediting care for the patients who need it most.
A powerful example comes from an AI tool integrated into chest X-ray workflows that automatically flags potential pneumothorax. In one hospital, this led to a significant reduction in time-to-treatment for patients with collapsed lungs cutting delays by over 50%. On the workflow side, AI-driven ultrasound guidance is under development to democratize access to this highly operator-dependent modality, which will enable less-experienced users to perform accurate scans, especially in rural or underserved areas.
Burnout in radiology is a growing concern, driven by rising imaging volumes and administrative tasks. AI can help reduce this burden in multiple ways: automating repetitive tasks, prioritizing urgent cases, and ensuring high-quality image acquisition at the point of care. Radiologists can then focus on what they do best: clinical reasoning and patient interaction. AI can also help support better work-life balance by streamlining workflows and reducing the cognitive load of decision-making.
We build AI solutions with and for clinicians. That means clinical co-development, robust validation in real-world environments, and continuous feedback loops. Our tools are FDA 510(k) cleared or CE-marked. Increasingly we are emphasizing explainability by designing our systems to provide visual overlays or confidence scores. We pay particular attention to testing across datasets that are representative of the intended population ensuring performance are maintained in various clinical contexts or across relevant population subgroups. AI must earn trust and that starts with transparency and clinical rigor.
One of AI’s most important contributions is standardization. It can help bring consistency to image interpretation, measurements, and reporting - independent of a radiologist’s experience or location. The goal is to reduce inter-reader variability, potentially minimize diagnostic errors, and ensure every patient receives a high standard of care.
Trust is earned, not assumed. That’s why we engage deeply with clinicians from the earliest stages of development. We prioritize education through programs like HelloAI and we ensure that our tools are integrated seamlessly into existing workflows, and designed to support, not supplant, clinical judgment. AI will not replace radiologists, but radiologists who use AI will be empowered to work smarter, faster, and with greater confidence. It’s about collaboration, not competition.
Radiology will become increasingly intelligent, interconnected, and patient-centered. AI will be embedded in every step of the imaging chain from acquisition to interpretation to reporting and follow-up. We’ll see more point-of-care diagnostics, real-time clinical decision support, and personalized imaging protocols. Radiologists will evolve into data-driven consultants who guide therapy and monitor outcomes, not just detect disease. The role is expanding, not diminishing, and AI is the catalyst for that transformation.
We must ensure fairness, safety, transparency, and accountability. That means addressing bias in training data, securing patient privacy, ensuring clinicians understand how decisions are made, and maintaining human oversight. Regulators are adapting, but the pace of innovation must be matched by thoughtful governance. We support global harmonization of standards and advocate for robust post-market surveillance to ensure continued safety and efficacy in clinical use.
What excites me most is the democratizing power of AI. We’re moving toward a world where high-quality diagnostics are no longer limited by geography or specialist availability. AI-enabled tools will help bring expert-level care to remote villages, mobile units, and even patients’ homes. Combined with cloud connectivity and digital ecosystems, this is not science fiction, it’s happening now. The potential to improve equity, access, and outcomes at scale is truly revolutionary.
Bias detection and mitigation are priorities. We source datasets from a variety of countries and clinical settings during training and validation to ensure representativeness of the device intended population. Our models are tested to confirm that accuracy is maintained across relevant subpopulations. Equity is essential to our mission of delivering better outcomes for all, not just some.
AI can extend the reach of radiology in places where specialists are scarce. For example, AI-guided portable ultrasound devices are enabling midwives and community health workers to perform basic exams and connect with remote experts. Cloud-based image-sharing and decision support are also bringing care closer to patients. It’s about shifting from centralized expertise to distributed intelligence and that’s a game-changer for global health equity.
Radiology will become a dynamic fusion of human expertise and intelligent technology - more proactive, precise, and personalized than ever before.
To the next generation of radiologists: embrace AI not as a threat, but as a tool that empowers you to be more impactful, more efficient, and more human in your care.
Prof. Dr. med. Mathias Goyen is the Chief Medical Officer for Imaging and Advanced Visualization Solutions at GE HealthCare. A radiologist by training and a passionate advocate for innovation, he focuses on bridging clinical insight with digital technology to advance global health equity and improve outcomes across healthcare systems.
