
Artificial intelligence is changing the way researchers work with medical imaging. What might take a doctor or researcher tens of minutes to examine manually can, in some cases, be analysed by AI in just a few seconds. More importantly, algorithms can help identify subtle patterns and structures that might otherwise be difficult to spot. This was one of the practical applications discussed during CLARA Days in Paris, where researchers and experts from different countries and disciplines came together to exchange knowledge and explore the growing role of AI in brain research and healthcare.
Finding what the human eye might miss
Medical imaging, particularly magnetic resonance imaging (MRI), provides researchers with vast amounts of information. Analysing hundreds of three-dimensional images and identifying small or complex lesions can be a demanding task. This is where AI can offer valuable support. Rather than focusing only on clearly visible structures, algorithms can help identify less obvious patterns across large sets of images.
"Some diagnostic methods still require a doctor to examine medical images visually, for example MRI scans, which can take tens of minutes. AI can analyse these images in just a few seconds, with comparable or potentially even better results," explains Milan Němý, a researcher at the Czech Institute of Informatics, Robotics and Cybernetics (CIIRC CTU) and Karolinska Institutet in Stockholm.
AI is particularly useful when researchers are looking for very small lesions, spatially complex structures or patterns that are difficult to distinguish visually. Its ability to process large volumes of data could help make image analysis faster and more precise.
Towards earlier detection of Alzheimer's disease
The potential applications go beyond improving the speed of image analysis. AI could also support efforts to detect neurodegenerative diseases, including Alzheimer's disease, at an earlier stage. Václav Snášel, CLARA Director, highlighted the importance of early detection and the role AI can play in supporting this process.
"Early detection of Alzheimer's disease helps patients. AI, in turn, helps us with this task," he explains. By identifying patterns in medical data and supporting the interpretation of complex images, AI-based methods could become an important tool for researchers and healthcare professionals working to better understand and detect neurodegenerative conditions.
Research needs more than algorithms
Developing AI applications for healthcare requires more than advanced algorithms. It also depends on access to computing infrastructure, strong academic research, education and collaboration between different sectors. These broader challenges were also part of the discussions in Paris.
Olivier Colliot(Paris Brain Institute) highlighted the importance of supporting education in mathematics and computer science, as well as creating the right conditions for research institutions, companies and students to contribute to AI development. His perspective underlined an important point: turning promising AI research into practical applications requires a strong ecosystem connecting scientific expertise, technology and investment.
Connecting expertise through CLARA
The discussions in Paris reflected the interdisciplinary and international nature of the CLARA project, which connects expertise in artificial intelligence, neuroscience and advanced computing. By bringing researchers together and creating opportunities to exchange knowledge across disciplines and institutions, CLARA contributes to an environment in which new approaches to understanding the brain and improving healthcare can develop.