While convolutional neural networks (CNNs) have been successfully applied to many challenging classification applications, they typically require large datasets for training. When the availability of labeled data is limited, data augmentation is a critical preprocessing step for CNNs. However, data augmentation for wearable sensor data has not been deeply investigated yet.In this paper, various data augmentation methods for wearable sensor data are proposed. The proposed methods and CNNs are applied to the classification of the motor state of Parkinson's Disease patients, which is challenging due to small dataset size, noisy labels, and large intra-class variability. Appropriate augmentation improves the classification performance from 77.54% to 86.88%.

IMPORTANCE Progressive supranuclear palsy (PSP) is a 4-repeat tauopathy. Region-specific tau aggregates establish the neuropathologic diagnosis of definite PSP post mortem. Future interventional trials against tau in PSP would strongly benefit from biomarkers that support diagnosis.OBJECTIVE To investigate the potential of the novel tau radiotracer 18 F-PI-2620 as a biomarker in patients with clinically diagnosed PSP. DESIGN, SETTING, AND PARTICIPANTSIn this cross-sectional study, participants underwent dynamic 18 F-PI-2620 positron emission tomography (PET) from 0 to 60 minutes after injection at 5 different centers (3 in Germany, 1 in the US, and 1 in Australia). Patients with PSP (including those with Richardson syndrome [RS]) according to Movement Disorder Society PSP criteria were examined together with healthy controls and controls with disease. Four additionally referred individuals with PSP-RS and 2 with PSP-non-RS were excluded from final data analysis owing to incomplete dynamic PET scans.

No abstract

No abstract

Parkinson's disease (PD) is a degenerative neurological disorder. Patients suffer from different movement disturbances. Amongst others, freezing of gait (FOG) is a common and disabling feature. A new method for FOG detection using a garment with integrated accelerometers, the MiMed-Pants, is presented and evaluated. MiMed-Pants are loose fitting washable pants for measuring movements and were previously developed. The accelerations of shanks and belt of 6 patients were recorded while walking on two standardized courses. The occurrence of FOG were detected and documented both by the new method and by a physician. Those recordings were then compared afterwards to determine the new method's sensitivity and specificity. With this wearable measurement device FOG could be detected without the distracting effects of an extensive measurement installation.