Publications

Background In Hereditary Spastic Paraplegia (HSP) type 4 (SPG4) a length-dependent axonal degeneration in the cortico-spinal tract leads to progressing symptoms of hyperrefexia, muscle weakness, and spasticity of lower extremities. Even before the manifestation of spastic gait, in the prodromal phase, axonal degeneration leads to subtle gait changes. These gait changes - depicted by digital gait recording - are related to disease severity in prodromal and early-to-moderate manifest SPG4 participants. Methods We hypothesize that dysfunctional neuro-muscular mechanisms such as hyperrefexia and muscle weak- ness explain these disease severity-related gait changes of prodromal and early-to-moderate manifest SPG4 partici- pants. We test our hypothesis in computer simulation with a neuro-muscular model of human walking. We introduce neuro-muscular dysfunction by gradually increasing sensory-motor refex sensitivity based on increased velocity feedback and gradually increasing muscle weakness by reducing maximum isometric force. Results By increasing hyperrefexia of plantarfexor and dorsifexor muscles, we found gradual muscular and kin- ematic changes in neuro-musculoskeletal simulations that are comparable to subtle gait changes found in prodromal SPG4 participants. Conclusions Predicting kinematic changes of prodromal and early-to-moderate manifest SPG4 participants by grad- ual alterations of sensory-motor refex sensitivity allows us to link gait as a directly accessible performance marker to emerging neuro-muscular changes for early therapeutic interventions. Keywords Gait simulation, Spasticity, Hyperrefexia, Prodromal, SPG4, HSP, Movement disorder

Previous functional imaging studies demonstrated body-selective patches in the primate visual temporal cortex, comparing activations to static bodies and static images of other categories. However, the use of static instead of dynamic displays of moving bodies may have underestimated the extent of the body patch network. Indeed, body dynamics provide information about action and emotion and may be processed in patches not activated by static images. Thus, to map with fMRI the full extent of the macaque body patch system in the visual temporal cortex, we employed dynamic displays of natural-acting monkey bodies, dynamic monkey faces, objects, and scrambled versions of these videos, all presented during fixation. We found nine body patches in the visual temporal cortex, starting posteriorly in the superior temporal sulcus (STS) and ending anteriorly in the temporal pole. Unlike for static images, body patches were present consistently in both the lower and upper banks of the STS. Overall, body patches showed a higher activation by dynamic displays than by matched static images, which, for identical stimulus displays, was less the case for the neighboring face patches. These data provide the groundwork for future single-unit recording studies to reveal the spatiotemporal features the neurons of these body patches encode. These fMRI findings suggest that dynamics have a stronger contribution to population responses in body than face patches.

In this chapter, methods will be introduced which are currently available to perform lesion-symptom mapping in patients with focal and degenerative cerebellar disease. At the beginning of the chapter, strength and weaknesses inherent in studies on localization of cerebellar function in these different patient populations will be discussed. Next, methods of lesion-symptom mapping in focal cerebellar disease will be explained in more detail including lesion delineation, lesion normalization, and descriptive and inferential statistical analysis. Finally, methods of lesion-symptom mapping in cerebellar degeneration and available atlases of the cerebellum in stereotaxic space will be introduced.

In this chapter, drug treatment in the few treatable causes of degenerative ataxias will first be presented including vitamin deficiency, metabolic disorders, and autoimmune disorders. Next, older and more recent drug trials will be discussed which have been performed in Friedreich’s ataxia and may lead to effective drug treatments in the future. Finally, it will be shown that, as yet, symptomatic treatments of most degenerative ataxias are lacking, that is, there is still no “anti-ataxic drug.” The only exceptions are aminopyridines and acetazolamide which can be beneficial in the treatment of downbeat nystagmus and episodic ataxias.

This ultrahigh field 7 T fMRI study addressed the question of whether there exists a core network of brain areas at the service of different aspects of body perception. Participants viewed naturalistic videos of monkey and human faces, bodies, and objects along with mosaic-scrambled videos for control of low-level features. Independent component analysis (ICA) based network analysis was conducted to find body and species modulations at both the voxel and the network levels. Among the body areas, the highest species selectivity was found in the middle frontal gyrus and amygdala. Two large-scale networks were highly selective to bodies, dominated by the lateral occipital cortex and right superior temporal sulcus (STS) respectively. The right STS network showed high species selectivity, and its significant human body-induced node connectivity was focused around the extrastriate body area (EBA), STS, temporoparietal junction (TPJ), premotor cortex, and inferior frontal gyrus (IFG). The human body-specific network discovered here may serve as a brain-wide internal model of the human body serving as an entry point for a variety of processes relying on body descriptions as part of their more specific categorization, action, or expression recognition functions.

When TMS is applied over motor cortex, it elicits movements that can be recorded in humans as motor-evoked muscle potentials, as well as in patterns in EEG. A discussion has been started recently in the community that TMS may not only excite neuronal structures in the central nervous system, but also cause peripheral co-stimulation of sensory and motor axons of the meninges, blood vessels, skin, and muscle. These structures may also excite the same cortical site that TMS was meant to stimulate in the first place, resulting in contamination of the TMS-induced cortical response. Therefore, many efforts are made to identify and isolate peripheral evoked potentials (PEPs) from TMS-induced cortical responses in EEG-Data. However, it is very difficult to develop an appropriate sham stimulation for humans that closely reflects auditory, somatosensory, and motor responses accompanying TMS. An obvious route to clarify the issue is the blockade of cranial nerves, which requires animal models where invasive experiments to discover putative areas of origin can be done. In recent years, we have developed a method to demonstrate the direct effect of a TMS pulse at the cellular level. We have transferred single pulse and repeated stimulation protocols from humans to a rat model. With selective blockade of PEP, we were able to show that the trigeminal nerve is a major contributor to TMS-evoked neuronal signals in motor cortex, represented by a prominent excitatory peak at around 20 ms after stimulation. TEPs starts much earlier and lasts up to 6 ms after the stimulus pulse. Both inputs then merge into a canonical inhibition-excitation pattern lasting more than 350 ms.

Summary The temporal cortex represents social stimuli, including bodies. We examine and compare the contributions of dynamic and static features to the single-unit responses to moving monkey bodies in and between a patch in the anterior dorsal bank of the superior temporal sulcus (dorsal patch [DP]) and patches in the anterior inferotemporal cortex (ventral patch [VP]), using fMRI guidance in macaques. The response to dynamics varies within both regions, being higher in DP. The dynamic body selectivity of VP neurons correlates with static features derived from convolutional neural networks and motion. DP neurons’ dynamic body selectivity is not predicted by static features but is dominated by motion. Whereas these data support the dominance of motion in the newly proposed “dynamic social perception” stream, they challenge the traditional view that distinguishes DP and VP processing in terms of motion versus static features, underscoring the role of inferotemporal neurons in representing body dynamics.

People can innately recognize human facial expressions in unnatural forms, such as when depicted on the unusual faces drawn in cartoons or when applied to an animal’s features. However, current machine learning algorithms struggle with out-of-domain transfer in facial expression recognition (FER). We propose a biologically-inspired mechanism for such transfer learning, which is based on norm-referenced encoding, where patterns are encoded in terms of difference vectors relative to a domain-specific reference vector. By incorporating domain-specific reference frames, we demonstrate high data efficiency in transfer learning across multiple domains. Our proposed architecture provides an explanation for how the human brain might innately recognize facial expressions on varying head shapes (humans, monkeys, and cartoon avatars) without extensive training. Norm-referenced encoding also allows the intensity of the expression to be read out directly from neural unit activity, similar to face-selective neurons in the brain. Our model achieves a classification accuracy of 92.15% on the FERG dataset with extreme data efficiency. We train our proposed mechanism with only 12 images, including a single image of each class (facial expression) and one image per domain (avatar). In comparison, the authors of the FERG dataset achieved a classification accuracy of 89.02% with their FaceExpr model, which was trained on 43,000 images.

Accurate real-time models of human motion are important for applications in areas such as cognitive science and robotics. Neural networks are often the favored choice, yet their generalization properties are limited, particularly on small data sets. This paper utilizes the Gaussian process dynamical model (GPDM) as an alternative. Despite their successes in various motion tasks, GPDMs face challenges like high computational complexity and the need for many hyperparameters. This work addresses these issues by integrating the information bottleneck (IB) framework with GPDMs. The IB approach aims to optimally balance data fit and generalization through measures of mutual information. Our technique uses IB variable selection as a component of GPLVM back-constraints to reduce parameter count and to select features for latent space optimization, resulting in improved model accuracy.

For social species, including primates, the recognition of dynamic body actions is crucial for survival. However, the detailed neural circuitry underlying this process is currently not well understood. In monkeys, body-selective patches in the visual temporal cortex may contribute to this processing. We propose a physiologically-inspired neural model of the visual recognition of body movements, which combines an existing image-computable model (`ShapeComp') that produces high-dimensional shape vectors of object silhouettes, with a neurodynamical model that encodes dynamic image sequences exploiting sequence-selective neural fields. The model successfully classifies videos of body silhouettes performing different actions. At the population level, the model reproduces characteristics of macaque single-unit responses from the rostral dorsal bank of the Superior Temporal Sulcus (Anterior Medial Upper Body (AMUB) patch). In the presence of time gaps in the stimulus videos, the predictions made by the model match the data from real neurons. The underlying neurodynamics can be analyzed by exploiting the framework of neural field dynamics.

Abstract. Wearable inertial measurement units (IMU) enable largescale multicenter studies of everyday gait analysis in patients with rare neurodegenerative diseases such as cerebellar ataxia. To date, the quantity of sensors used in such studies has involved a trade-off between data quality and clinical feasibility. Here, we apply machine learning techniques to potentially reduce the number of sensors required for real-life gait analysis from three sensors to a single sensor on the hip. We trained 1D-CNNs on constrained walking data from individuals with cerebellar ataxia and healthy controls to generate synthetic foot data and predict gait features from a single sensor and tested them in free walking conditions, including the everyday life of unseen subjects. We compare 14 stride-based gait features (e.g. stride length) with three sensors (two on the feet and one on the hip) with our approach estimating the same features based on raw IMU-data from a single sensor placed on the hip. Leveraging layer-wise relevance propagation (LRP) and transfer learning, we determine driving elements of the input signals to predict individuals’ gait features. Our approach achieved a relative error (

Since neural networks play an increasingly important role in critical sectors, explaining network predictions has become a key research topic. Counterfactual explanations can help to understand why classifier models decide for particular class assignments and, moreover, how the respective input samples would have to be modified such that the class prediction changes. Previous approaches mainly focus on image and tabular data. In this work we propose SPARCE, a generative adversarial network (GAN) architecture that generates SPARse Counterfactual Explanations for multivariate time series. Our approach provides a custom sparsity layer and regularizes the counterfactual loss function in terms of similarity, sparsity, and smoothness of trajectories. We evaluate our approach on real-world human motion datasets as well as a synthetic time series interpretability benchmark. Although we make significantly sparser modifications than other approaches, we achieve comparable or better performance on all metrics. Moreover, we demonstrate that our approach predominantly modifies salient time steps and features, leaving non-salient inputs untouched.

Background: Telemedicine interventions support behavioral state-of-the-art treatment of obsessive-compulsive disorder (OCD) as therapy can be delivered in the patients' home environment, allowing for more ecologically valid symptom actualization and access to experts even in rural areas. Sensors to indicate a patient´s emotional state and gaze direction during exposures with response prevention help to adapt therapy individually and to prevent avoidance behavior. This study will investigate the feasibility and acceptability of sensor-based telemedical treatment for children with OCD in the home setting. Methods: We plan to develop the therapy system with 10 healthy children and 5-10 children with OCD, aged 12-18 years, and then to evaluate it by treating 20 children with OCD of the same age group in 14 weekly therapy sessions via teleconference. We will use eye trackers to record the patient´s gaze and pupillometry, while the heart rate is captured by an ECG chest belt to identify stress responses. Inertial sensors capture movements to detect behavioral patterns. An app is used to record the children's self-rated symptoms and emotional state on a daily basis. Pre- and post-study questionnaires on obsessive-compulsive symptoms, feasibility and acceptance of the therapy by children, parents and therapists will be evaluated. Conclusion: We expect this therapeutic approach to show good feasibility and significant symptom reduction, as well as improvement for psychotherapeutic interventions through direct feedback of physiological responses within therapy sessions. We will further explore the underlying mechanisms in OCD treatment before applying them to other disorders.