The Department of Machine Learning for Clinical Neuroscience is currently recruiting PhD candidates and Postdocs. We develop advanced machine and deep learning models to analyze diverse clinical data, including neuroimaging, psychometric, clinical, smartphone, and omics datasets. While focusing on methodological challenges (explainability, robustness, multimodal data integration, causality etc.), the main goal is to enhance early diagnosis, predict disease progression, and personalize treatment for neurological and psychiatric diseases in diverse clinical settings. We offer an exciting and supportive environment with access to state-of-the-art compute facilities, mentoring and career advice through experienced faculty. Hertie AI closely collaborates with the world-class AI ecosystem in Tübingen (e.g. Cyber Valley, Cluster of Excellence “Machine Learning in Science”, Tübingen AI Center).

Neurodegenerative diseases involve accumulation of aberrant proteins in the brain, leading to brain damage and progressive cognitive and behavioral dysfunction. Many gaps exist in our understanding of how these diseases initiate and how they progress through the brain. However, evidence has accumulated supporting the hypothesis that aberrant proteins can be transported using the brain’s intrinsic network architecture — in other words, using the brain’s natural communication pathways. This theory forms the basis of connectome-based computational models, which combine real human data and theoretical disease mechanisms to simulate the progression of neurodegenerative diseases through the brain. In this talk, I will first review work leading to the development of connectome-based models, and work from my lab and others that have used these models to test hypothetical modes of disease progression. Second, I will discuss the future and potential of connectome-based models to achieve clinically useful individual-level predictions, as well as to generate novel biological insights into disease progression. Along the way, I will highlight recent work by my lab and others that is already moving the needle toward these lofty goals.

Microglia are the resident CNS macrophages of the brain parenchyma. They have many and opposing roles in health and disease, ranging from inflammatory to anti-inflammatory and protective functions, depending on the developmental stage and the disease context. In Multiple Sclerosis, microglia are involved to important hallmarks of the disease, such as inflammation, demyelination, axonal damage and remyelination, however the exact mechanisms controlling their transformation towards a protective or devastating phenotype during the disease progression remains largely unknown until now. We wish to understand how brain microglia respond to demyelinating insults and how their behaviour changes in recovery. To do so we developed a novel histopathological analysis approach in 3D and a cell-based analysis tool that when applied in the cuprizone model of demyelination revealed region- and disease- dependent changes in microglial dynamics in the brain grey matter during demyelination and remyelination. We now use similar approaches with the aim to unravel sensitive changes in microglial dynamics during neuroinflammation in the EAE model. Furthermore, we employ constitutive knockout and tamoxifen-inducible gene-targeting approaches, immunological techniques, genetics and bioinformatics and currently seek to clarify the specific role of the brain resident microglial NF-κB molecular pathway versus other tissue macrophages in EAE.

Ophthalmology is ideally placed to benefit from recent advances in artificial intelligence. It is a highly image-based specialty and provides unique access to the microvascular circulation and the central nervous system. This talk will demonstrate diverse applications of machine learning and deep learning techniques in ophthalmology, including in age-related macular degeneration (AMD), the leading cause of blindness in industrialized countries, and cataract, the leading cause of blindness worldwide. This will include deep learning approaches to automated diagnosis, quantitative severity classification, and prognostic prediction of disease progression, both from images alone and accompanied by demographic and genetic information. The approaches discussed will include deep feature extraction, label transfer, and multi-modal, multi-task training. Cluster analysis, an unsupervised machine learning approach to data classification, will be demonstrated by its application to geographic atrophy in AMD, including exploration of genotype-phenotype relationships. Finally, mediation analysis will be discussed, with the aim of dissecting complex relationships between AMD disease features, genotype, and progression.

How seizures emerge from the abnormal dynamics of neural networks within the epileptogenic tissue remains an enigma. Are seizures random events, or do detectable changes in brain dynamics precede them? Are mechanisms of seizure emergence identical at the onset and later stages of epilepsy? Is the risk of seizure occurrence stable, or does it change over time? A myriad of questions about seizure genesis remains to be answered to understand the core principles governing seizure genesis. The last decade has brought unprecedented insights into the complex nature of seizure emergence. It is now believed that seizure onset represents the product of the interactions between the process of a transition to seizure, long-term fluctuations in seizure susceptibility, epileptogenesis, and disease progression. During the lecture, we will review the latest observations about mechanisms of ictogenesis operating at multiple temporal scales. We will show how the latest observations contribute to the formation of a comprehensive theory of seizure genesis, and challenge the traditional perspectives on ictogenesis. Finally, we will discuss how combining conventional approaches with computational modeling, modern techniques of in vivo imaging, and genetic manipulation open prospects for exploration of yet hidden mechanisms of seizure genesis.

Circadian clocks dominate our lives. By creating and distributing an internal representation of 24-hour solar time, they prepare us, and thereby adapt us, to the daily and seasonal world. Jet-lag is an obvious indicator of what can go wrong when such adaptation is disrupted acutely. More seriously, the growing prevalence of rotational shift-work which runs counter to our circadian life, is a significant chronic challenge to health, presenting as increased incidence of systemic conditions such as metabolic and cardiovascular disease. Added to this, circadian and sleep disturbances are a recognised feature of various neurological and psychiatric conditions, and in some cases may contribute to disease progression. The “head ganglion” of the circadian system is the suprachiasmatic nucleus (SCN) of the hypothalamus. It synchronises the, literally, innumerable cellular clocks across the body, to each other and to solar time. Isolated in organotypic slice culture, it can maintain precise, high-amplitude circadian cycles of neural activity, effectively, indefinitely, just as it does in vivo. How is this achieved: how does this clock in a dish work? This presentation will consider SCN time-keeping at the level of molecular feedback loops, neuropeptidergic networks and neuron-astrocyte interactions.

There are currently no approved therapies to slow down the accumulation of neurological disability that occurs independently of relapses in multiple sclerosis (MS). International agencies are engaging to expedite the development of novel strategies capable of modifying disease progression, abrogating persistent CNS inflammation, and support degenerating axons in people with progressive MS. Understanding why regeneration fails in the progressive MS brain and developing new regenerative approaches is a key priority for the Pluchino Lab. In particular, we aim to elucidate how the immune system, in particular its cells called myeloid cells, affects brain structure and function under normal healthy conditions and in disease. Our objective is to find how myeloid cells communicate with the central nervous system and affect tissue healing and functional recovery by stimulating mechanisms of brain plasticity mechanisms such as the generation of new nerve cells and the reduction of scar formation. Applying combination of state-of-the-art omic technologies, and molecular approaches to study murine and human disease models of inflammation and neurodegeneration, we aim to develop experimental molecular medicines, including those with stem cells and gene therapy vectors, which slow down the accumulation of irreversible disabilities and improve functional recovery after progressive multiple sclerosis, stroke and traumatic injuries. By understanding the mechanisms of intercellular (neuro-immune) signalling, diseases of the brain and spinal cord may be treated more effectively, and significant neuroprotection may be achieved with new tailored molecular therapeutics.

Protein quality control is essential for maintenance of a healthy and functional proteome that can attend the multiplicity of cellular functions. Failure of the systems that contribute to protein homeostasis, the so called proteostasis networks, have been identified in the pathogenesis of multiple neurodegenerative disorders and demonstrated to contribute to disease onset and progression. We are interested in autophagy, one of the components of the proteostasis network, and in the interplay of wo selective types of autophagy, chaperone-mediated autophagy (CMA) and endosomal microautophagy (eMI), with neurodegeneration. We have recently found that pathogenic proteins involved in common neurodegenerative conditions such as tauopathies or Parkinson’s disease, can exert a toxic effect in both types of selective types of autophagy compromising their functioning. We have now used mouse models with compromised CMA that support increased propagation of proteins such as tau and alpha-synuclein and an exacerbation of disease phenotype with aging. Conversely, genetic or chemical upregulation of CMA in this context of proteotoxicity slow down disease progression by facilitating effective intracellular removal of pathogenic proteins. Our findings highlight CMA and eMI as potential novel therapeutic targets against neurodegeneration.

The normally functioning blood-brain barrier (BBB) regulates the transfer of material between blood and brain. BBB dysfunction has long been recognized in multiple sclerosis (MS), and there is considerable interest in quantifying functional aspects of brain blood vessels and their role in disease progression. Parenchymal water content and its association with volume regulation is important for proper brain function, and is one of the key roles of the BBB. There is convincing evidence that the astrocyte is critical in establishing and maintaining a functional BBB and providing metabolic support to neurons. Increasing evidence suggests that functional interactions between endothelia, pericytes, astrocytes, and neurons, collectively known as the neurovascular unit, contribute to brain water regulation, capillary blood volume and flow, BBB permeability, and are responsive to metabolic demands. Increasing evidence suggests altered metabolism in MS brain which may contribute to reduced neuro-repair and increased neurodegeneration. Metabolically relevant biomarkers may provide sensitive readouts of brain tissue at risk of degeneration, and magnetic resonance offers substantial promise in this regard. Dynamic contrast enhanced MRI combined with appropriate pharmacokinetic modeling allows quantification of distinct features of BBB including permeabilities to contrast agent and water, with rate constants that differ by six orders of magnitude. Mapping of these rate constants provides unique biological aspects of brain vasculature relevant to MS.

Mindsets are core assumptions about the nature and workings of things in the world that orient us to a particular set of attributions, expectations, and goals. Our study of mindsets is, in part, inspired by research on the placebo effect, a robust demonstration of the ability of mindsets, conscious or subconscious, to elicit physiological changes in the body. This talk will explore the role of mindsets in three stages of chronic disease progression: genetic predisposition, behavioral prevention, and clinical treatment. I will discuss the mechanisms through which mindsets influence health as well as the myriad ways that mindsets can be more effectively leveraged to motivate healthy behaviors and improve 21st century healthcare.

The spatial patterning of each neurodegenerative disease relates closely to a distinct structural and functional network in the human brain. This talk will mainly describe how brain network-sensitive neuroimaging methods such as resting-state fMRI and diffusion MRI can shed light on brain network dysfunctions associated with pathology and cognitive decline from preclinical to clinical dementia. I will first present our findings from two independent datasets on how amyloid and cerebrovascular pathology influence brain functional networks cross-sectionally and longitudinally in individuals with mild cognitive impairment and dementia. Evidence on longitudinal functional network organizational changes in healthy older adults and the influence of APOE genotype will be presented. In the second part, I will describe our work on how different pathology influences brain structural network and white matter microstructure. I will also touch on some new data on how brain network integrity contributes to behavior and disease progression using multivariate or machine learning approaches. These findings underscore the importance of studying selective brain network vulnerability instead of individual region and longitudinal design. Further developed with machine learning approaches, multimodal network-specific imaging signatures will help reveal disease mechanisms and facilitate early detection, prognosis and treatment search of neuropsychiatric disorders.

Cross-sectional and longitudinal multimodal brain imaging studies using positron emission tomography (PET) and magnetic resonance imaging (MRI) have provided detailed insight into the pathophysiological progression of Alzheimer’s disease. It starts at an asymptomatic stage with widespread gradual accumulation of beta-amyloid and spread of pathological tau deposits. Subsequently changes of functional connectivity and glucose metabolism associated with mild cognitive impairment and brain atrophy may develop. However, the rate of progression to a symptomatic stage and ultimately dementia varies considerably between individuals. Mathematical models have been developed to describe disease progression, which may be used to identify markers that determine the current stage and likely rate of progression. Both are very important to improve the efficacy of clinical trials. In this lecture, I will provide an overview on current research and future perspectives in this area.