By Doreen Witt
06 May 2026

Interview with Dr Tobias Lindig

New Alzheimer's therapies: Why modern image management and AI determine success

With the approval of antibody drugs such as Lecanemab and Donanemab, a new era is dawning in the treatment of Alzheimer's: for the first time, the progression of the disease can be slowed down. However, this medical breakthrough brings with it massive logistical challenges. Alzheimer's therapies require extremely early diagnosis and regular, close MRI monitoring. The consequence: increasing numbers of scans and more complex diagnoses. Without modern Alzheimer's therapy image management, seamless data exchange and artificial intelligence (AI), this new form of care can hardly be implemented across the board. Dr Tobias Lindig, senior physician at the University Hospital of Tübingen and co-founder of AIRAmed, explains what this means in practice.

Medical imaging is becoming a key factor in the early detection of Alzheimer's and other neurodegenerative diseases. This is because it relies heavily on high-precision MRI scans, as subtle structural changes in the brain often occur years before the first clinical symptoms appear.

In order to identify suitable patients for the new treatment options with Lecanemab and Donanemab and to safely monitor side effects, this image data must not only be generated in enormous quantities, but must also be made available, compared and securely managed across all institutions. Efficient image data management therefore forms the backbone of modern dementia care.

We discussed why imaging plays such a central role in early detection and how radiologists can cope with the growing flood of data with a recognised expert on this topic: Dr Tobias Lindig, senior physician at the University Hospital of Tübingen and co-founder of the AI start-up AIRAmed explains in an interview how AI-supported brain volumetry works as a „gatekeeper“ and why IT providers must work hand in hand to provide the best possible technical support for diagnostics.

1 Dr Lindig, you are co-founder of a start-up for AI-supported brain volumetry and, as a senior physician for diagnostic and interventional neuroradiology, you have many years of experience in the early detection of Alzheimer's disease. Can you tell us why you founded the company on this topic?

Together with my co-founder and longstanding colleague Prof Benjamin Bender, I have been working scientifically on neurodegenerative diseases for over 20 years, in particular on the question of how Alzheimer's and other dementias can be detected as early as possible. We both work at the University Hospital in Tübingen and see every day how great the diagnostic uncertainties are, especially in the early stages of these diseases.

We realised relatively early on that although classic imaging techniques are extremely valuable, they reach their visual limits in the early stages of the disease. Many relevant changes in the brain are so subtle that they cannot yet be reliably recognised with the naked eye.

AIRAmed was created precisely on the basis of this clinical and scientific experience. AIRAmed develops AI-based software for the early detection of dementia, Alzheimer's and related neurodegenerative diseases. Our technology analyses MRI image data of the brain, measures the volumes of individual brain regions and systematically relates these to reference values of healthy test subjects. This enables us to objectively classify whether the brain volume is age-appropriate or whether pathological changes are already present that indicate a neurodegenerative disease.

This approach is particularly crucial in the early stages: small structural changes can often not yet be reliably assessed visually, but can be visualised with the help of quantitative, AI-supported analyses. Our aim is therefore not to replace medical findings, but to complement them in a meaningful way and provide radiologists and neurologists with an additional, objective basis for decision-making.

Dr Tobias Lindig

Co-founder & Managing Director of AIRAmed GmbH, Tübingen (since 2019)
Senior Physician in the Department of Diagnostic and Interventional Neuroradiology at the University Hospital Tübingen
Medical qualification: Specialist in neurology & radiology,
Specialising in quantitative imaging and clinical studies, completed medical studies in Erlangen, Boston, Zurich and Auckland

2. With the antibody drugs Lecanemab and Donanemab, disease-modifying therapies for Alzheimer's are available for the first time. How do you assess this breakthrough from a medical and economic perspective - is this the turning point that many have been waiting for?

From a medical point of view, Lecanemab and Donanemab really are a real breakthrough. They are the first therapies in around 20 years that have been proven to have a disease-modifying effect and not just treat symptoms. They therefore mark a turning point in Alzheimer's therapy.

However, it is important to categorise them correctly: these drugs do not cure Alzheimer's, they slow down the progression of the disease - especially if they are used very early on. This is precisely their great opportunity, but also their greatest challenge.

This opens up several opportunities for doctors: for the first time, we have a therapeutic option that influences the course of the disease. This changes the entire diagnostic logic: early detection becomes not only desirable, but medically relevant. It is also to be expected that these two substances - similar to MS therapies around 20 years ago - will only mark the beginning and that further, improved active substances will follow.

In my view, the risks are sometimes unnecessarily dramatised in the public debate. The known side effects, especially ARIA changes, are real, but well known, monitorable and manageable.

This new phase also brings economic challenges: the therapies are cost-intensive and require structured diagnostics, close imaging and interdisciplinary collaboration. At the same time, however, there is also enormous potential - for patients through longer independence and quality of life, and for healthcare systems in the long term through potentially lower care and follow-up costs.

To summarise: Yes, from my perspective, this is the beginning of a new era in Alzheimer's treatment. The decisive factor will be whether we manage to create the necessary diagnostic, financial and regulatory framework conditions so that these therapies can be used responsibly, early and fairly.

3. The new treatment options are only effective at an early stage. What diagnostic procedures are necessary to identify suitable patients?

When we talk about early detection in the context of new therapies, I believe that structural imaging is the ideal initial step in dementia assessment - especially in the form of AI-based brain volumetry.

A quantitative MRI analysis can provide an early answer: Are there already pathological changes? Is further clarification necessary at all? And if so, in which direction should it go? This is precisely what is clinically crucial in order to provide patients with targeted further management.

It should not be forgotten that clinical symptoms usually only appear when relevant parts of the brain have already been damaged. Neuropsychological tests therefore often come too late or are too unspecific at a very early stage to reliably differentiate between normal ageing and the onset of neurodegeneration.

Brain volumetry can take on a gatekeeper function here: It enables early, objective classification and ensures that complex resources such as amyloid PET or cerebrospinal fluid diagnostics are used specifically as confirmatory procedures - for example, when the volume pattern already clearly indicates a pathology typical of Alzheimer's disease.

4 Where do you currently see the biggest hurdles from a medical and technological perspective?

Of course, it is not always easy to get an MRI appointment promptly in Germany, especially in the statutory health insurance sector. But you can't fool yourself: There is no alternative that would be easier to access.

It is therefore a particular challenge to utilise the available resources in an efficient, structured and evidence-based manner and to integrate the medical and technological complexity of the entire topic into care in a meaningful way. If this succeeds, early detection can be transformed from a theoretical aspiration into a practical reality.

5 A central aspect of Alzheimer's therapy is the monitoring of ARIA - the side effects of the treatment. What demands does this place on radiologists?

The monitoring of ARIA is a central component of therapy with amyloid antibodies, but also places new, very specific demands on radiology. Above all, it involves structured longitudinal monitoring with clearly defined time points and questions.

Radiologists must be able to reliably recognise ARIA, classify it correctly and evaluate it in a clinical context. This requires experience with these hitherto rather rare image findings, a high sensitivity for subtle signal changes and the reliable differentiation between ARIA-E, ARIA-H and other age- or vascular-related changes, in particular the differentiation from an acute stroke.

ARIA: The most important side effect of anti-amyloid therapy

ARIA (Amyloid-Related Imaging Abnormalities) refers to characteristic MRI changes under antibody therapies for Alzheimer's disease. A distinction is made between ARIA-E (oedema/effusions) and ARIA-H (cerebral microbleeds and superficial siderosis). Most cases are asymptomatic and resolve spontaneously, although headaches, confusion or neurological symptoms may occur in rare cases. Close MRI monitoring is therefore essential.

6 In your opinion, are the radiology departments prepared to reliably recognise these specific side effects?

In part - but certainly not across the board. ARIA has not yet been an everyday topic in routine care. Accordingly, there is a clear need for additional qualifications, standardised protocols and close coordination with neurology and memory outpatient clinics. Radiology is becoming more than ever a part of active therapy management for neurodegenerative diseases, similar to oncology or chronic inflammatory CNS diseases.

Technologically, standardised MRI protocols, structured reporting and quantitative, AI-supported analyses can help to make findings more objective and reproducible and detect subtle changes in the course with greater certainty.

In short: ARIA monitoring makes it clear that we not only need new drugs in Alzheimer's therapy, but also further development of radiological expertise. Radiologists are thus becoming central players in a new, complex therapy concept - and must be prepared accordingly.

7 The new therapies also require complex monitoring. How do you assess the technological prerequisites for implementing the therapies across the board in Germany?

I see the biggest challenge less in technical data management than in the organisational dimension. If around ten per cent of Alzheimer's sufferers qualify for Alzheimer's treatment in the future, this will mean above all a relevant increase in additional MRI examinations - not just once, but serially as part of diagnostics, therapy control and ARIA monitoring.

It is precisely this quantity of scans that is the real hurdle for many radiological facilities: the availability of appointments, standardised protocols, prompt diagnosis and reproducible progress comparisons must function under already high equipment and personnel utilisation.

A decisive efficiency lever therefore lies in the imaging itself. Targeted sequence optimisation on the part of the manufacturer allows examinations to be carried out much more efficiently. In addition, AI- and LLM-based diagnostic assistants can support radiologists with structured analyses and progression comparisons.

8. Can artificial intelligence not only help with basic diagnostics, but also with recognising ARIA side effects earlier and more reliably?

This is actually an area in which I have had to significantly revise my own assessment of recent years. If you had asked me two years ago, I would probably have said: AI is not needed to detect ARIA - any radiologist can see microbleeds and oedema straight away.

Today I realise that this was too short-sighted. Of course, there are clear cases of ARIA that are easily recognisable visually. In practice, however, there are also many borderline and mixed findings: very small microbleeds, small oedema or changes that are difficult to distinguish from age-related or vascular findings - especially during the course of the disease.

This is precisely where artificial intelligence can provide real added value. AI-supported analyses help to detect subtle changes more quickly, objectively quantify differences in progression and reproducibly compare findings. This increases safety - especially in therapy monitoring, where early and reliable detection of side effects is crucial.

9 How do different players - doctors, but also software and IT providers - need to work together so that these therapies can be successfully integrated into standard care?

There needs to be close coordination between all those involved along a clear care pathway. Treating physicians are responsible for the indication, treatment decision and clinical follow-up. Radiology plays a central role in structured basic diagnostics, quantitative evaluation and ARIA monitoring throughout the course of treatment.

IT and platform providers such as Telepaxx enable seamless access to preliminary examinations, follow-up data and second opinions across different institutions. AI providers such as AIRAmed support radiologists with objective quantification, standardised evaluations and consistent progress comparisons - especially with increasing numbers of examinations.

Practical standards at all levels are crucial for success: In medical terms, this requires clearly defined MRI protocols, standardised diagnostic logic and fixed times for follow-up checks. Technologically, the consistent use of established standards such as DICOM forms the indispensable basis for loss-free and cross-system data exchange. Only when these medical and technical guidelines interlock seamlessly in a functioning digital network will the therapeutic breakthrough become a scalable and realistically realisable care solution in everyday clinical practice.

Conclusion: networking as a success factor

The new Alzheimer's therapies mark a medical milestone - but their successful implementation in standard care will largely depend on the technological infrastructure. As Dr Lindig makes clear in the interview, it is not just about individual innovations, but about the intelligent interaction of all those involved: referring physicians and neurologists who diagnose and treat at an early stage. Radiologists who monitor in a structured manner and reliably recognise ARIA risks. AI solutions that enable objective quantification and progression comparisons. And platform providers that ensure the seamless, standardised exchange of image data between all stakeholders. Only when these pieces of the puzzle fit together can the therapeutic breakthrough become a scalable healthcare reality - for the benefit of patients and their families.