In bipedal gait, the initiation of the first step is preceded by a complex sequence of movements which shift the centre of mass of the body towards the stance foot to allow for a step of the swing foot. These anticipatory postural adjustments (APAs) have been investigated in order to elucidate movement strategies in healthy and diseased persons. We studied the influence of several external parameters (age, type of step initiation) on APAs and investigated whether Parkinsonian patients may have different APAs. As a result, we found that externally elicited steps were preceded by faster and larger APAs than self-timed steps. Parkinsonian patients without the freezing of gait (FOG) phenomenon showed overall slightly reduced APAs but did not clearly differ from patients with FOG. Multiple APAs were seen in up to 25 % of the steps of the patients and in a much lower percentage of the steps of control subjects. The results indicate that APAs are significantly influenced by the timing of a step, i.e. are larger in externally elicited steps. The patients showed an overall preserved APA pattern but slowed movements and amplitude, indicating that increased bradykinesia due to progressive illness is a plausible explanation for these findings. The freezing phenomenon is not explained by a general absence or massive reduction in APA measures.

PD patients with implanted DBS electrodes allow for the investigation of basal ganglia activity during gait. Whereas the overall spectral content of local field potentials (LFP) in the BG is known to change with movement, it is still unclear how exactly the BG-LFP change during gait. We investigated the electric local field potentials of the STN of 10 patients with PD during gait. Gait parameters were recorded with inertial sensor units and the gait was subsequently analyzed in parallel to the electrophysiological activity in Matlab. Additionally, we recorded LFPs during rest and control experiments to probe for movement related artifacts. Stimulation was off in all experiments reported here. On a group level, a frequency analysis shows that the high beta frequency band gets significantly downregulated during gait as compared to sitting rest. On an individual level, a linear classifier trained with frequency profiles of small epochs of LFP activity recorded during gait and rest can distinguish between walking and resting episodes with a mean classification accuracy across all patients of 90%. A time–frequency analysis on the group level reveals a distinct modulation pattern. Whereas low frequencies in the theta and alpha range and high frequencies in the gamma range are modulated up and down time-locked to the gait cycle, the beta band gets modulated down during the complete gait cycle.

PD patients with implanted DBS electrodes allow for the investigation of basal ganglia activity during gait. Whereas the overall spectral content of local field potentials (LFP) in the BG is known to change with movement, it is still unclear how exactly the BG-LFP change during gait. We investigated the electric local field potentials of the STN of 10 patients with PD during gait. Gait parameters were recorded with inertial sensor units and the gait was subsequently analyzed in parallel to the electrophysiological activity in Matlab. Additionally, we recorded LFPs during rest and control experiments to probe for movement related artifacts. Stimulation was off in all experiments reported here. On a group level, a frequency analysis shows that the high beta frequency band gets significantly downregulated during gait as compared to sitting rest. On an individual level, a linear classifier trained with frequency profiles of small epochs of LFP activity recorded during gait and rest can distinguish between walking and resting episodes with a mean classification accuracy across all patients of 90%. A time–frequency analysis on the group level reveals a distinct modulation pattern. Whereas low frequencies in the theta and alpha range and high frequencies in the gamma range are modulated up and down time-locked to the gait cycle, the beta band gets modulated down during the complete gait cycle.

Inertial sensors are electric circuits which measure acceleration and rotational velocity. This information is needed to determine the 3-D position of a body in space. With the help of those sensors gait can be recorded. The method was validated by a motion-capture system. We examined twelve healthy young men. Sensors were attached at the thigh as well as at the lower leg on each side. One sensor consisted of one gyroscope and two accelerometers. With the help of an integrated Kalman Filter it was possible to calculate the position of all 4 leg segments in the sagittal plane. Specific pattern of gyroscopic and accelerometric data helped defining moments of terminal contact (toe-off) and initial contact (heel strike). These events were validated by a pressure-sensitive inserted sole. To reconstruct the position of the swing foot, forward kinematics were used, i.e. the stance foot was taken a pivotal point and the position of the leg segments was computed with reference to that point so it was possible to calculate step length and step velocity. We compared this data with data synchronously obtained from a motion capture system while subjects walked on a treadmill. Comparison of the time of initial and terminal contact obtained by inertial sensors and pressure sensitive soles resulted in very small deviations within the range of data resolution. Correlation between segment angles measured with inertial sensors and motion capture system was above r = 0.9. Correlation between step length measured by these two systems was above r > 0.8 and could be increased by a calibration of the inertial-sensors system by a defined gait distance. Gait analysis with inertial sensors is less expensive than with conventional methods and allows for the assessment of gait in more natural environments. The actual disadvantage of less precision is compensated in future with better sensors and software.

Arm swing asymmetry during gait may be a sensitive sign for early Parkinson's disease. There is only very limited information about how much asymmetry can be considered to be physiological. To assess the normal range of arm swing asymmetry, we investigated 60 healthy subjects. The influence of age, gender, and additional mental tasks (dual-tasking) on arm swing asymmetry was assessed. Limb kinematics of 60 healthy persons in three age groups (between 40 and 75 years) were measured with an ultrasound motion capture system while subjects walked on a treadmill. Treadmill velocity was varied (3 steps) and mental loads (2 different tasks) were applied in different trials. Additionally, a group of 7 patients with early Parkinson's disease was investigated. Arm swing amplitude as well as arm swing asymmetry varied considerably in the healthy subjects. Elderly subjects swung their arms more than younger participants. Only the more demanding mental load caused a significant asymmetry, i.e., arm swing was reduced on the right side. In the patient group, asymmetry was considerably higher and even more enhanced by mental loads. Our data indicate that an asymmetry index above 50 (i.e., one side has twice the amplitude of the other) may be considered abnormal. Evaluation of arm swing asymmetry may be used as part of a test battery for early Parkinson's disease. Such testing may become even more important when disease-modifying drugs become available for Parkinson's disease.

Problems during gait constitute a severe disability for patients with Parkinson’s disease (PD) and lead to significant limitations during their daily life, especially in later stages of the disease. The nuclei of the basal ganglia are crucial for important aspects of motor processes and are used as target areas for deep brain stimulation therapy of the motor symptoms associated with PD. Investigations of the (patho)-physiology of these structures during gait should therefore contribute to our understanding of the neuronal processes involved during normal movement and clarify if symptom-specific pathological neuronal activity can be characterized. For this purpose, we investigated the electric local field potentials of the subthalamic nucleus of 12 patients with PD recorded from deep brain stimulation electrodes during walking and during rest. Simultaneously, the movement of the patients was recorded with inertial sensor units and subsequently analyzed in parallel to the electrophysiological activity. In this way, gait patterns like normal walking or akinetic freezing epochs can be reconstructed and their specific neuronal activity can be described with the analysis of the electrophysiological recordings. Attenuation of oscillations in the beta frequency band (12–35 Hz) can be observed in patients without freezing epochs during normal walking as compared to rest. In contrast, patients with freezing epochs show an increase in lower beta frequency (12–22 Hz) amplitude during walking as compared to rest as well as freezing epochs themselves show increased lower beta frequency as compared to normal walking and rest. Both, patients with nearly regular gait and patients with akinetic epochs did show an increase in the gamma frequency band (35–90 Hz) during movement. Our results show that an increase of oscillations in the lower beta band is highly correlated with severe freezing epochs and might possibly play a causal role for akinesia in PD. Such a symptom-specific and characteristic electrophysiological fingerprint could be used as a kind of marker of pathological neuronal activity for the adjustment of the deep brain stimulation therapy in a closed-loop fashion.

S OF THE 7 TH POSTURE SYMPOSIUM

Deep Brain Stimulation (DBS) is an established surgical treatment for neurodegenerative diseases such as Parkinson’s Disease (PD) or Dystonia. In terms of imaging, pre-operative MRI scans are used for surgical planning and post-operative CT scans for control of electrode placement. Intra-operatively, however, imaging guidance is often avoided, due to contra-indications of DBS electrode contacts with MRI and difficulty of acquiring CT scans. Ultrasound imaging could be an attractive alternative, but has not been investigated yet due to difficult positioning of the transducer in the narrow burr-holes. Recently, there has been interest in the community towards the usage of non-invasive transcranial ultrasound (TCUS) through the pre-aureal bone window. In a seminal work (Walter, 2012 ), Walter described the intra-operative usage of TCUS for lead electrode imaging and post-operative placement assessment of DBS electrode, showing its potential and limitations. In this work, we investigate, for the first time, the feasibility of using 3D-TCUS for imaging of already implanted Deep Brain Stimulation (DBS) electrodes. As a first step towards this goal, we report electrode localization errors outside of the operating room on six previously operated DBS patients. Post-operative CT scans are registered to already acquired pre-operative T1 and T2 MRI sequences to localize the actual electrode tip locations in MRI. We then acquire 3D-TCUS sweeps and co-register them to the pre-operative MRI as well, using a head-surface point-cloud registration first (using Iterative Closest Points, ICP) and a manual refinement using several pairs of corresponding points on anatomical structures (midbrain, ventricles) in TCUS and MRI. The comparison between electrode tips identified in 3D-TCUS and actual tip positions in CT are a first step towards assessing whether 3D-TCUS could provide a benefit intra-operatively in future. Quantitative evaluation is performed by reporting the accuracy of electrode tip localization after ICP-based registration (mean 5.62 mm, stdev 2.26 mm) and after manual registration refinement (mean 4.09 mm, stdev 1.75 mm). Additionally, we provide multiple image examples showing the dependence of 3D-TCUS quality depending on the bone window, and multiple examples of image artifacts in TCUS images and 3D reconstructions. Intra-operative 3D ultrasound imaging has potential for electrode tip localization during DBS procedures. 3D transcranial US could be an attractive alternative to intra-cranial imaging through the narrow burr-hole. We report the first results of volumetric 3D-TCUS imaging on already operated patients in a post-operative scenario. Initial results still show a relatively high inaccuracy of electrode tip localization, e.g. due to TCUS artifacts and imperfect registration to the pre-operative MRI. Nevertheless, the potential of TCUS for DBS electrode imaging as described by Walter (2012) can be extended to 3D, allowing for electrode tip and shaft localization. This motivates further work, e.g. on improved MRI-US registration and computer-aided tip detection (Fig. 1, Fig. 2).