Magnetic resonance imaging (MRI) can now visualize both the structure and function of the brain in extraordinary detail. But how can we identify subtle changes among thousands of data points that precede the development of Alzheimer’s or other neurodegenerative diseases?

Artificial intelligence is entering medicine in ways that would have seemed like science fiction only a few years ago. This is precisely the approach used by Milan Němý from CIIRC CTU in his search for answers. In his research, he combines advanced imaging methods with AI tools capable of detecting patterns in data that are invisible to the human eye.

On the occasion of Brain Awareness Week, we bring you an interview about how brain scans can be “read” in a new way and what this could mean for the future of early diagnosis.

At the end of 2025, Milan Němý was selected by an international committee as a Tenure Track Position Holder within the European center of excellence CLARA Centre (Center for Artificial Intelligence and Quantum Computing in System Brain Research). The center focuses on linking artificial intelligence, bioinformatics, and brain research.

His work is therefore gaining further institutional support for an ambitious goal: to better understand how the brain changes in the earliest stages of disease and how these changes can be detected reliably and in time.

You work at the intersection of AI, brain imaging, and neuroscience. How would you describe your work to someone outside the scientific community? What real-world problems are you trying to solve?

My main goal is to understand how the brain changes long before visible symptoms appear in diseases such as Alzheimer’s. We use modern MRI techniques and artificial intelligence to capture extremely subtle changes in brain structure and connectivity, changes that are often impossible to detect with the naked eye.

Simply put, we are trying to teach computers to “read” brain MRI scans more sensitively and more precisely than humans can. Our aim is to detect vulnerability early, understand which brain systems are affected, and predict how the disease may evolve over time and across different parts of the brain.

If we can recognize risk earlier and more reliably, it opens the door to prevention and personalized care.

Many brain diseases, including Alzheimer’s disease and brain tumors, are still diagnosed relatively late. What key challenge does your research address, and how does AI help? Why is MRI such a powerful tool in this context?

The fundamental problem is that brain diseases often develop gradually and quietly. When the first symptoms appear, significant and often irreversible damage has already occurred in the brain.

MRI is exceptional because it allows us to look noninvasively into the living brain and quantify its structure and function in great detail. Today, however, we no longer see MRI simply as a tool for “taking pictures” of brain anatomy.

In the context of our research, it acts more like a physicochemical analyzer that captures microscopic properties of tissues as well as spatial information about brain networks. This makes it possible to detect early signals of disease that are subtle and dispersed, rather than appearing as clear, visible lesions.

To analyze large MRI datasets across multiple sequences, together with other clinical information, artificial intelligence is essential. Machine learning models can identify combinations of features and trajectories of change that would be practically impossible for humans to detect manually.

Instead of asking, “Is this brain already damaged?” we can begin to ask, “Is this brain at risk? And how is it changing over time?”

A significant part of your work focuses on very subtle brain changes, for example in the cholinergic system in Alzheimer’s disease. What have we been missing so far, and what new insights are you gaining with these methods?

Research on Alzheimer’s disease has long focused on classic markers such as amyloid, tau, or hippocampal atrophy. These are important, but they do not fully explain why the brain becomes vulnerable in the earliest stages.

Much less attention has been given to the fact that early changes may affect specific brain networks, such as those involved in attention, memory modulation, or cognitive flexibility, typically the cholinergic system. If this system is disrupted, it can affect a wide range of functions.

The problem is that these pathways are spatially complex, and for a long time it was not possible to measure them precisely and sensitively using conventional imaging methods.

In our team, we have developed methods for segmenting the cholinergic system in the brain. When we combine this with diffusion MRI and AI-supported analysis, we can quantify its integrity directly in living people, something that was previously impossible.

This opens a new direction in research. Instead of studying the late consequences of disease, we can focus more on its beginnings. It also allows us to study younger individuals at a stage when pathological processes are just beginning, but classical methods still “see nothing.”

You recently obtained a new tenure-track position at the CLARA Centre of Excellence in bioinformatics and AI. What attracted you to CLARA, and how does this position support your scientific independence and long-term vision?

What attracted me to CLARA is that it brings together experts from many different fields within a single framework: artificial intelligence, high-performance computing, quantum technologies, and neuroscience.

This kind of synergy is essential if we want to address complex problems such as neurodegeneration, which cannot be solved within a single discipline. Even though this may sound obvious, in practice it is still not the standard in science. That is why I greatly value what the CLARA Centre is building.

From a practical perspective, the tenure-track position gives me scientific independence to develop my own research direction while also allowing me to be part of a collaborative environment with strong technological and computational infrastructure.

I believe that the combination of independence and a well-designed ecosystem is ideal for building a stable long-term research team and program.

What are your main goals at the CLARA Centre in the coming years? What are you most looking forward to?

My main ambition is to build a strong interdisciplinary team connecting neuroscientists, AI specialists, radiologists, and neurologists. I want to create a group that is scientifically independent, sustainable in the long term, and capable of pushing the boundaries of neuroimaging and computational methods.

From a scientific perspective, one priority is the further validation and expansion of our framework for assessing the cholinergic system. We are currently developing it toward PET imaging, longitudinal analyses of large cohorts, and integration into clinical drug trials.

For each of these areas, we already have international partners, and we look forward to testing our methods in different populations and real clinical environments.

A second key area is associated pathologies, especially cerebrovascular disease. These conditions significantly contribute to cognitive decline, yet robust and standardized tools for describing the spatial distribution and characteristics of vascular lesions in the brain are still lacking.

We are therefore developing new, technologically advanced software that combines advanced imaging and machine learning to describe vascular brain damage more systematically and in a biologically meaningful way. Without strong expertise in AI and imaging, this would not be possible.

Our goal is to make this tool available to the research community and potentially, in the future, to clinical practice.

It is also important to note that most of these activities rely on international collaborations. These partnerships provide not only complementary expertise but also access to new datasets, without which AI models cannot be properly validated or generalized.

Neurodegenerative diseases are among the greatest challenges for aging societies. What role do you think AI-based brain research could play in how these diseases are diagnosed or treated in the future?

I believe AI can genuinely change the situation. Even today, it is clear that it can accelerate research and reveal patterns and relationships that we previously could not recognize.

Ideally, AI could help uncover the fundamental principles of these diseases so that one day we might move toward effective prevention. It is difficult to say exactly when such a breakthrough might happen, but we have seen similar shifts in other fields.

In the near term, however, I expect more practical impacts. The combination of AI and biomarkers such as MRI, PET, and others will allow us to identify people at risk much earlier than today.

This could lead to targeted monitoring, personalized prevention strategies, and better timing of therapeutic interventions. It will also improve clinical trials for new drugs, for example by selecting suitable participants and using more sensitive outcome measures.

Overall, this represents a step toward precision neurology, where decisions are based on the biology of an individual brain rather than broad diagnostic categories.

Ing. Milan Němý, Ph.D.

CIIRC CTU

Milan Němý leads a neuroscience research group at the Czech Institute of Informatics, Robotics and Cybernetics at CTU in Prague (CIIRC CTU), within the Department of Cognitive Systems and Neuroscience (COGSYS). He is also a researcher at Karolinska Institutet in Sweden, one of the world’s leading medical universities and the institution responsible for awarding the Nobel Prize in Physiology or Medicine.

His work focuses on neurodegenerative diseases, particularly Alzheimer’s disease, and on the use of advanced MRI methods combined with artificial intelligence. His research contributes to a better understanding of early brain changes and to the development of more sensitive biomarkers for early diagnosis.

He has received several prestigious awards, including the Jan Bureš Award 2023 for the best scientific article in neuroscience, the GAČR Junior Star grant in 2024 to establish his own research team, and the Junior Faculty Award at the AD/PD™ 2025 international conference.

At CIIRC CTU, he holds a tenure-track position co-funded by the European Center of Excellence CLARA, where he develops research at the intersection of artificial intelligence, bioinformatics, and neuroscience to improve early diagnosis and understanding of brain diseases.